Mindfulness (2022) 13:449–461 https://doi.org/10.1007/s12671-021-01806-2 ORIGINAL PAPER Differential Brain Perfusion Changes Following Two Mind–Body Interventions for Fibromyalgia Patients: an Arterial Spin Labelling fMRI Study Sonia Medina1 · Owen G. O’Daly1 · Matthew A. Howard1 · Albert Feliu‑Soler2 · Juan V. Luciano2,3 Accepted: 4 December 2021 / Published online: 5 January 2022 © Crown 2021 Abstract Objectives Further mechanistic insight on mind–body techniques for fibromyalgia (FMS) is needed. Arterial spin labelling (ASL) imaging can capture changes in regional cerebral blood flow (rCBF) that relate to spontaneous pain. Methods We recruited FMS patients undergoing either mindfulness-based stress reduction training (MBSR, n = 14) or a psychoeducational programme (FibroQoL, n = 18), and a control FMS group with no add-on treatment (n = 14). We acquired whole-brain rCBF maps and self-report measures at baseline and following treatment and explored interaction effects in brain perfusion between the treatment group and session with a focus on the amygdala, the insula and the anterior cingulate cortex (ACC). Results We identified a significant interaction effect in the amygdala, which corresponded with rCBF decreases following FibroQoL specifically. At baseline, rCBF in the amygdala for the FibroQoL group correlated with pain catastrophizing and anxiety scores, but not after treatment, suggesting a decoupling between activity in the amygdala and negative emotional symptoms of FMS as a consequence of treatment. Baseline rCBF correlated positively with pain symptoms in the ACC and the anterior insula across all patients; moreover, the correlation between rCBF changes post intervention in the insula and pain improvement was negative for both treatments and significantly different from the control group. We suggest that there is disruption of the typical relationship between clinical pain and activity as a product of these two nonpharmacological therapies. Conclusions We have demonstrated that different mind-to-body treatments correspond to differential changes in clinical symptoms and brain activity patterns, which encourages future research investigating predictors of treatment response. Trial Registration NCT02561416. Keywords Fibromyalgia · Mindfulness · Meditation, pCASL fMRI Fibromyalgia syndrome (FMS) is a chronic condition char- heterogeneity of early symptoms and unknown primary acterized by widespread musculoskeletal pain, fatigue, cause of this condition, FMS diagnosis is very often by stiffness, sleep disturbances, cognitive problems and dis- exclusion, long after onset or following prior misdiagnoses tress (Häuser et al., 2015). Because of a lack of consist- (Wolfe et al., 2019). Given the broad spectrum of symp- ency of diagnostic criteria (Goldenberg, 2009), the high toms present in FMS, it is crucial to adopt a multidiscipli- nary approach to manage FMS at all stages of the disor- der (Nüesch et al., 2013). Despite this, standard treatments * Sonia Medina available have traditionally only targeted alleviation of pain Sonia.medina@kcl.ac.uk symptoms with only modest success, namely tricyclic anti- 1 Department of Neuroimaging, King’s College London, depressants, opioids, sedatives, tramadol and nonsteroidal London, UK anti-inflammatory drugs (Sumpton & Moulin, 2008). More 2 Department of Clinical & Health Psychology, Autonomous recently, nonpharmacological interventions have started to University of Barcelona, Bellaterra, Spain be incorporated as part of standard clinical care, including 3 Research & Innovation Unit, Parc Sanitari Sant Joan de Déu, physical therapy (Stucki, 2000) or cognitive behavioural Sant Boi de Llobregat, Spain therapy (Bennett & Nelson, 2006), mindfulness (Aman 13 Vol.:(0123456789) 450 Mindfulness (2022) 13:449–461 et al., 2018; Pérez-Aranda, D’Amico, et al., 2019), virtual study how rCBF relates to FMS symptom improvement reality therapy (Garcia-Palacios et al., 2015) or clinical hyp- as well as mechanisms of action of treatments for FMS, nosis (Picard et al., 2013). However, little-to-no knowledge the literature in this field is to date scarce (Moana-Filho exists regarding their mechanisms of action within the cen- et al., 2013; Müller et al., 2021; Zeidan et al., 2018). A tral nervous system. Accordingly, to date, biomarkers that comprehensive model of the mechanism of action of novel facilitate treatment response prediction have yet to be devel- non-pharmacological techniques for FMS using perfusion oped (Häuser et al., 2015) in order to promote individualized fMRI is yet to be established. treatment plans for patients. The present study includes data from a subsample of FMS Neuroimaging techniques are useful tools to study mech- patients who took part in the EUDAIMON study (Pérez- anisms of action of nonpharmacological interventions for Aranda, D’Amico, et al., 2019; Pérez-Aranda, Feliu-Soler, clinical pain (Jensen et al., 2012a, b, c), specifically studies et al., 2019), a randomized, controlled trial (RCT) that employing functional magnetic resonance imaging (fMRI) assessed the efficacy and cost-utility of mind–body interven- using blood oxygen level-dependent (BOLD) contrasts. tions to manage FMS symptoms: mindfulness-based stress They have provided evidence of altered functional con- reduction (MBSR), and a psychoeducational and relaxation nectivity (FC) of the default mode network (DMN) (Fallon training programme (FibroQoL) (Luciano, Aguado, et al., et al., 2016) and cognitive control network in FMS patients 2013; Luciano, Sabes-Figuera, et al., 2013). Both interven- compared to healthy controls (Kong et al., 2018), as well tions are accepted as active treatments; they are equivalent as changes in the aforementioned networks following tran- in structure, produce similar reductions in FMS impact on scranial direct current stimulation to the motor cortex (Cum- patients’ daily lives in the long term (Pérez-Aranda, Feliu- miford et al., 2016), cognitive behavioural therapy (Jensen Soler, et al., 2019) and demonstrate superior cost-utility than et al., 2012a, b, c), hypnosis (Derbyshire et al., 2009) or treatment as usual alone; however, whether the neurophysi- mindfulness (Young et al., 2018). The amygdala seems to ological mechanisms of action of both treatments are compa- play a key role in encoding major targets of psychologi- rable is yet to be assessed. We applied ASL fMRI to acquire cal therapies for chronic pain, such as depression (Giesecke rCBF maps from two groups of FMS patients prior to and et al., 2005) or anxiety and fear of pain (Hsiao et al., 2020). following one of these interventions in addition to their treat- Studies using sustained-pain BOLD fMRI paradigms point ment as usual (TAU). An additional control group of FMS towards FC increases of the somatosensory cortex with patients who continued with TAU only was also recruited. the insula during pain in FMS patients (Kim et al., 2015) We sought to determine (a) whether rCBF changed differ- which can be altered with cognitive therapy (Lazaridou entially between treatment groups, in order to better under- et al., 2017); they also suggest that there is less FC within stand the mechanism of action of each treatment; (b) how the pain inhibitory network during pressure pain, using the rCBF changes relate to FMS symptom improvements and rostral anterior cingulate cortex (ACC) as seed (Jensen et al., (c) the utility of rCBF as a viable biomarker of clinical pain 2012a, b, c). The subgenual ACC (sgACC) is an important in FMS. We hypothesized that rCBF in the amygdala would hub for descending pain control and analgesia (Schrepf et al., reduce following treatment, as a consequence of reduced 2016; Sprenger et al., 2011), and has previously shown defi- stress and improved anxiety control (Jiang et al., 2016); fur- cient activation during pain tasks in FMS patients (Jensen thermore, since mind–body interventions such as mindful- et al., 2009). Moreover, the insula and the ACC are directly ness training aim to improve cognitive control, emotional linked via the salience network, which is essential for cog- regulation and body awareness, we hypothesized that MBSR nitive control and holds a complementary function with and FibroQoL would induce changes within the salience the DMN (Jilka et al., 2014). Nevertheless, as Jensen et al., network via perfusion reductions at rest in insula and ACC (2012a, b, c) suggested, results from traditional fMRI stud- following treatment. ies relate to experimental paradigms optimized in the lab, where BOLD fMRI techniques rely on rapid, higher fre- quency fluctuations in hemodynamic signals. BOLD fMRI Methods is therefore not sufficiently sensitive to capture low-fre- quency, long-lasting fluctuations more closely related to Participants spontaneous persistent pain and natural clinical improve- ment. Arterial spin labelling (ASL) is an fMRI technique A total of 90 FMS female patients recruited in this RCT that offers an alternative to BOLD via measuring cerebral participated in the neuroimaging sub-study. Due to image blood flow (rCBF) across the brain, and is capable of meas- artefacts and missing imaging data in one of the study ses- uring those slow subtle perfusion differences (Loggia et al., sions and participant withdrawal, we included data from a 2019). Although ASL presents a valuable opportunity to total of only 46 patients in the final analysis. Participants 13 Mindfulness (2022) 13:449–461 451 were recruited from a database of 531 patients at the Rheu- allocation. In the next 5 days, all patients from the neuro- matology Service at Parc Sanitari Sant Joan de Déu, St. imaging substudy were contacted to complete a baseline Boi de Llobregat (Spain). Patients had been previously MRI scanning session. Following their first scanning ses- diagnosed with FMS according to the American College sion, participants then underwent their allocated study of Rheumatology 1990 Criteria at the recruitment site. intervention for the following 8 weeks; once the treat- Inclusion criteria were the following: (i) female, right- ment phase was finalized, participants attended a follow- handed patients between 18 and 65 years of age, (ii) able to up visit, where an identical clinical evaluation and MRI understand Spanish language, (iii) willing to provide writ- scanning were carried out (for a flowchart of the study ten consent to participate. Exclusion criteria included the design, see study protocol (Feliu-Soler et al., 2016) and following: (i) participation in other RCTs, (ii) presence of Fig. 1 and 2). cognitive impairment according to the Mini-Mental State The following interventions were administered: Examination (MMSE < 27), (iii) receiving psychological treatment during the previous or current year, (iv) previous Intervention Group 1 (TAU + MBSR) Participants underwent experience in meditation or other mind–body therapies, the MBSR protocol developed at the University of Mas- (v) comorbid mental disorder or severe medical illness sachusetts Medical School, USA, consisting of 8 weekly which may interfere with the study, (vi) inability to attend 2-h-long group sessions (n = 15) of mindfulness training. group sessions, (vii) pregnancy, (viii) involvement in ongo- MBSR was developed to help people with chronic pain and ing litigation relating to FMS. Additional exclusion crite- other stress-related conditions (Kabat-Zinn, 1982). To avoid ria for the neuroimaging substudy participation included exhaustion or withdrawal from the study, session duration (i) usual MRI contraindications (claustrophobia, metal was slightly shortened (2 instead of 2.5 h). Additionally, we implants, pacemakers, etc.), (ii) consuming more than 8 offered patients an optional half-day of silent MBSR retreat caffeine units per day (1 caffeine drink was permitted on between weeks 6 and 7. Both the group and the homework the day of the study, (iii) smoking more than 5 cigarettes sessions included elements of MBSR practice such as body per day, (iv) acute pain not related to FMS on the day of scan, sitting meditation and mindful movements/stretches, the study (e.g. headache, lumbar pain). Participants were with the purpose of helping patients to relate to their physi- instructed to refrain from taking any rescue analgesic aside cal and psychological symptoms in more accepting and non- from TAU 72 h prior to each MRI session. Since patients judgmental ways. Patients were provided with a MBSR book were also evaluated for their immune-inflammatory sta- (Martin-Ausero, 2008). tus (results published in Andrés-Rodríguez et al. 2019), additional exclusion criteria were also taken into account Intervention Group 2 (TAU + FibroQoL) FibroQoL was struc- for the present study: infection/cold symptoms on the day turally equivalent to MBSR (8 weekly 2-h-long sessions) of blood extraction; needle phobia; BMI > 36 kg/m 2 or and consisted of a group-based (n = 15) psycho-educational weight > 110 kg; neoplastic illnesses, infection, cardiopul- programme specific to FMS patients, led by specialized cli- monary, vascular, or other internal conditions (collected nicians, divided into two parts: 4 sessions of psychoeduca- from the medical history); use of oral or local corticoster- tion in which patients receive information about the patho- oids or anticytokine therapy. physiology, diagnosis and management of FM symptoms and another 4 sessions of training in self-hypnosis to gener- Procedure ate a state of deep relaxation with the objective of achieving control over the body and pain (a focus that is inconsistent The present study was part of a 12-month, parallel-group, with mindfulness) and to imagine one’s life in the future randomized, single-blind, controlled trial with three treat- without pain. ment arms, TAU + MBSR, TAU + FibroQoL, and TAU Patients allocated in the active treatment arms were only. The Consolidated Standards of Reporting Trials instructed to continue their usual pattern of medication and 2010 (CONSORT; Schulz et al., 2010) were followed. recorded their usual care over the course of the RCT. Audio- Participants underwent an initial clinical evaluation to tapes were also provided to both groups to facilitate practice confirm diagnosis and eligibility. Eligible participants at home. were randomly allocated to one of the three treatment groups (i.e. TAU, TAU + MBSR, and TAU + FibroQoL), Treatment as Usual Control Group (TAU) The TAU provided and from there, they were assigned to either the main was mainly pharmacological as per usual Spanish clinical RCT or the RCT + neuroimaging substudy. Participants practice, accompanied by counselling on aerobic exercise attended a baseline clinical evaluation blind to treatment according to individual patients’ physical conditions. 13 452 Mindfulness (2022) 13:449–461 Fig. 1  Flowchart of experimental design for the EUDAIMON neuroimaging substudy. Figure adapted from EUDAIMON study protocol Fig. 2  Results from mixed 3 by 2 ANOVA and post hoc pairwise comparisons. Initial cluster-forming threshold = 0.005. Results are significant at peak level after family-wise error correction at alpha = 0.05 post SVC in the amygdala 13 Mindfulness (2022) 13:449–461 453 Measures parameters estimated from the normalization step of the T2 scans as well as the study-specific template. Finally, we The measures listed below were the ones included in the multiplied each normalized functional image by a binary, neuroimaging sub-study by the researchers, as they may con- skull-stripped T2 template in order to remove any extracer- tribute to explain inter-individual variability in the imag- ebral tissue from the images. We performed manual quality ing data results. For a detailed summary of all the outcome assurance at each step to identify artefacts (e.g. co-registra- measures, the reader is referred to the study protocol (Feliu- tion failures) as well as outliers. Halfway through MRI data Soler et al., 2016). acquisition an unexpected, unplanned update of the scanner software took place, causing the functional images acquired Demographic and Clinical Measures We collected the fol- afterwards to be scaled differently. To mitigate this issue, lowing information from participants: age, years from FMS we calculated the mean global rCBF for each image and diagnosis, pharmacological medication as part of their TAU divided each voxel by their respective mean global rCBF and psychiatric comorbidities (e.g. depression). value, resulting in normalized, comparable rCBF maps. Self‑report Clinical Measures In all clinical evaluations (i.e. Data Analyses baseline and follow-up sessions), participants completed a series of standardized questionnaires, in order to better char- Clinical Data Once we checked that normality assumption acterize patients’ phenotypes and to assess changes in symp- was met, we compared baseline data across the three treat- toms associated with FMS after treatment. These were (i) the ment groups via a one-way analysis of variance (ANOVA) Revised Fibromyalgia Impact Questionnaire (FIQR) (Ben- with subsequent post hoc tests in order to ensure the groups nett et al., 2009; Luciano, Aguado, et al., 2013; Luciano, were comparable. We also explored changes in each one Sabes-Figuera, et al., 2013), (ii) the Hospital Anxiety and of the self-report measures of interest following treatment Depression Scale (HADS) (Luciano et al., 2014; Zigmond & via paired samples t-test. We performed these analyses in Snaith, 1983) and (iii) the Pain Catastrophizing Scale (PCS) SPSS v. 26. (García-Campayo et al., 2010; Sullivan et al., 1995). Partici- pants also rated their perceived pain intensity on the day of Neuroimaging Data We adopted a mass univariate general each scanning session using a paper-and-pencil horizontal, linear model approach for all group-wise statistical analy- 100-mm visual analogue scale anchored with ‘no pain’ and ses of the pCASL data. We set all initial cluster-forming ‘maximum pain’ (Langley & Sheppeard, 1985). height thresholds to p < 0.005 and applied family-wise error (FWE) correction at cluster extent p < 0.05. For each one MRI Data Acquisition and Preprocessing We performed MRI of our regions of interest, which we examined via a small measurements on a 3.0-T Phillips Ingenia wide-bore MR volume correction (SVC) for each contrast, we applied scanner, equipped with an 8-channel, phased-array, receive- FWE correction at cluster peak level. The ROIs chosen for only head coil. All patients had an axial T2-weighted 3D analyses were the amygdala bilaterally, the insula bilaterally, turbo spin-echo (TSE) pulse sequence with slice thick- the pregenual ACC and the sgACC. We chose the insula ness = 3 mm, repetition time = 3000 ms, echo time = 95 ms, and amygdala predefined masks from the WFU_PickAtlas flip angle = 90° and field of view = 240 × 180 × 125 mm. Toolbox in SPM. The pregenual and sgACC ROIs were For brain perfusion measurements, we used a fast, single- extracted from the Brainnetome Atlas Parcellation Toolbox shot, pCASL sequence (repetition time = 5000 ms, label (Fan et al., 2016). We explored treatment-induced changes distance = 90 mm, label duration = 1650 ms, post-labelling in rCBF via a mixed 3 × 2 ANOVA with ‘Treatment’ as delay = 1600 ms, voxel size = 1.875 × 1.875 × 3 mm). We between-subjects factor (i.e. TAU + MBSR, TAU + Fibro- finally quantified a regional cerebral blood flow (rCBF) map QoL and TAU only) and ‘Period’ as within-subjects factor per patient for each MRI session. For image preprocessing, (i.e. baseline and follow-up). We included patients’ ages to we used the Statistical Parametric Mapping software (SPM) the model as an additional nuisance covariate. We examined version 12. First, we segmented the structural T2 images and the Treatment arm × Period interaction effect and explored normalized the resulting grey matter (Zigmond and Snaith) the results further via post hoc paired t-tests (i.e. baseline masks to Montreal Neurological Institute (MNI) space using vs follow-up scans) within each treatment group. We fur- non-linear registration tools from the DARTEL toolbox in ther investigated rCBF changes observed in the amygdala SPM. For this step, we created an intermediate study-specific by extracting the mean rCBF value of the amygdala (across template to avoid bias associated with group normalization. the entire ROI) and calculating Pearson’s correlations with We then performed co-registration of the pCASL images to self-report measures in SPSS v. 26. We also examined the native-space T2 scans. The co-registered functional images correlation between mean rCBF in the amygdala at baseline were then warped into MNI space using the deformation and changes on self-report clinical measures after treatment, 13 454 Mindfulness (2022) 13:449–461 as well as correlation between baseline self-report measures to these results, we sought to determine whether changes in and delta rCBF scores in the amygdala (i.e. baseline minus rCBF in the TAU + FibroQoL group in the amygdala related follow-up). For effect size estimation, Hedge’s g statistics to a modulation of negative affective comorbid symptoms were calculated as a function of Cohen’s d in order to avoid present in chronic pain, as it has been previously suggested inflation effects due to sample sizes below 20 (Hedges & (Neugebauer et al., 2004). Accordingly, we computed Pear- Olkin, 2014). Finally, we examined the relationship between son’s correlations between amygdala mean rCBF and both self-report clinical scores and rCBF across our ROIs by the HADS anxiety subscale and the PCS (Fig. 3). At base- means of a multiple regression analysis in SPM. In this case, line, mean rCBF in the amygdala correlated negatively with a separate model was estimated for each one of the behav- HADS anxiety scores (r =  − 0.57, p = 0.028) and positively ioural datasets, using clinical measures scores at baseline with PCS total scores (r = 0.56, p = 0.016). Post hoc multiple and delta scores (i.e. baseline minus follow-up) as regres- regression analysis combining the three PCS subscales (i.e. sors. We estimated additional regression models including rumination, magnification and helplessness) revealed that all participants and a separate intercept per treatment group, baseline rCBF could not be predicted by a linear combina- in order to look at whether regression slopes (i.e. linear rela- tion of the subscales (overall model fit R2 = 0.408, p = 0.07) tionship between changes in rCBF and changes in clinical and only the rumination subscale alone at baseline was a scores) differed across groups. significant predictor of amygdala rCBF (overall model fit R2 = 0.305, p = 0.007; ­CBFamygdala = 0.628 × Rumina- tion + 0.877). None of these correlations reached signifi- Results cance for the follow-up assessment. We found no significant correlation between rCBF in the amygdala and VAS scores Self‑report Clinical Measures in either period. In order to further interpret these results, we investigated whether baseline rCBF in the amygdala for this The final sample size was n = 14 for the TAU + MBSR treatment group could predict changes in PCS and HADS group, n = 18 for the TAU​+ FibroQoL group and n = 14 for anxiety scores after treatment (i.e. delta scores) and vice the TAU only group. At baseline, there were not significant versa. Baseline rCBF correlated positively with delta PCS differences across the three treatment groups in terms of scores (i.e. P ­ CSbaseline − ­PCSfollow up), that is, greater rCBF age, years from FMS diagnosis, and FIQR, HADS anxiety, levels at baseline corresponded with greater improvement of PCS and VAS scores. Paired t-test results indicated that the pain catastrophising (r = 0.56, p = 0.018). Baseline rCBF did TAU + MBSR group experienced a reduction in PCS scores not correlate with delta HADS anxiety measures; however, for the magnification subscale after treatment (t(13) = 2.787, baseline HADS anxiety scores were positively correlated sig = 0.018); however, this result did not survive Bonfer- with delta rCBF (r = 0.71, p = 0.001), that is, greater anxi- roni’s correction for multiple comparisons across all PCS ety scores corresponded with greater rCBF decreases after subscales. The TAU​+ FibroQoL group presented score TAU + FibroQoL treatment. We observed no significant reductions in both the rumination and helplessness sub- results for baseline VAS scores or delta VAS scores. scales of the PCS (t(17) = 3.559, sig = 0.003 and t(17) = 2.964, sig = 0.009 respectively). There were no significant changes Regression Analyses in reported pain, FIQR or HADS measures following treat- ment across groups. We observed a reduction in reported We investigated how baseline rCBF may relate to any of our pain on the control group (t(13) = 3.067, sig = 0.009). A sum- self-report measures of interest via multiple linear regression mary of all results on self-report measures can be found in models. We found that baseline VAS scores could predict Table 1. baseline rCBF levels in the left AI (T = 4.20, pFWE = 0.026, peak coordinates (MNI, x y z) =  − 36 9 18, 286 voxels) and Effects of Treatment on rCBF the left sgACC (T = 3.96, pFWE = 0.013, peak coordinates (MNI, x y z) =  − 9 34 − 10, 112 voxels) across all partici- We examined changes in rCBF after treatment across all pants (Fig. 4). Baseline rCBF did not predict changes in treatment groups. Hypothesis-led analysis revealed a signifi- self-report measures after treatment in any of the patient cant interaction effect between ‘Treatment’ and ‘Period’ in groups; however, the regression slopes for the relationship the left amygdala (F = 9.26, pFWE = 0.037, peak coordinates between delta VAS scores and delta rCBF were significantly (MNI, x y z) =  − 26 − 4 − 18, 54 voxels). Post hoc pairwise different between treatment groups and the control group comparisons within each treatment group indicated that in the right AI (T = 3.68, p(FWE) = 0.043, peak coordinates this interaction effect was driven by a rCBF decrease after (MNI, x y z) = 44 10 − 6, 85 voxels) where there was a neg- TAU + FibroQoL treatment (T = 3.74, pFWE = 0.022, peak ative relationship between delta VAS and delta rCBF for coordinates (MNI, x y z) =  − 27 − 2 − 21, 58 voxels). Further the TAU + MBSR and TAU + FibroQoL treatments but a 13 Mindfulness (2022) 13:449–461 455 Table 1  Summary of self-report clinical data and results from paired t-test analyses across treatment groups TAU + MBSR TAU + MBSR TAU + FibroQoL TAU + FibroQoL TAU​ TAU​ Baseline (N = 14) Follow-up (N = 14) Baseline (N = 18) Follow-up (N = 18) Baseline (N = 14) Follow-up (N = 14) Age 54.15 (± 8.83) - 52.27 (± 8.68) - 51.5 (± 8.40) - Years from FMS 15.54 (± 9.97) - 9.76 (± 4.02) - 12.45 (± 11) - duration m.v. = 2 m.v. = 1 m.v. = 3 Mean VAS (SD) 5.74 (± 2.32) 4.99 (± 2.12) 6.04 (± 2.03) 5.61 (± 2.4) 6.84 (± 1.88) 5.15 (± 2.23) t (sig) 0.953 (0.361) 1.023 (0.321) 3.067 (0.009) Mean FIQR (SD) 61.97 (± 18.02) 48.79 (± 21.42) 65.75 (± 17.20) 58.82 (± 21.74) 59.304 (± 27.23) 59.54 (± 24.57) m.v. = 1 m.v. = 1 t (sig) 2.115 (0.056) 1.964 (0.067) − 0.097 (0.924) Mean HADS 7.46 (± 4.87) 7.07 (± 4.66) 9 (± 4.66) 7.35 (± 5.43) 8.21 (± 5.10) 9.38 (± 4.5) depression (SD) m.v. = 1 t (sig) 0.256 (0.802) 1.679 (0.113) − 0.705 (0.494) Mean HADS anxi- 9.92 (± 3.72) 9.30 (± 5.64) 12 (± 4.34) 9.94 (± 4.69) 9.14 (± 4.5) 10.46 (± 3.4) ety (SD) m.v. = 1 t (sig) 0.387 (0.706) 1.667 (0.115) − 1.474 (0.166) Mean PCS total 18.66 (± 15.06) 10.92 (± 8.21) 25.58 (± 14.02) 17.82 (± 13.63) 16.29 (± 10.78) 22.30 (± 14.91) (SD) m.v. = 1 t (sig) 2.825 (0.015) 2.901 (0.010) − 1.366 (0.197) Mean PCS rumina- 5.16 (± 4.7) 2.84 (± 2.67) 9.41 (± 4.67) 6.35 (± 4.51) 5.85 (± 4.58) 7.15 (± 4.77) tion (SD) m.v. = 1 m.v. = 1 t (sig) 2.092 (0.06) 3.559 (0.003) − 0.801 (0.439) Mean PCS magnifi- 4.75 (± 3.59) 2.69 (± 2.21) 4.94 (± 3.49) 3.88 (± 2.75) 3.64 (± 2.23) 5.07 (± 3.68) cation (SD) m.v. = 1 m.v. = 1 t (sig) 2.787 (0.018) 1.394 (0.182) − 1.921 (0.079) Mean PCS help- 8.75 (± 7.67) 5.38 (± 4.87) 11.23 (± 6.95) 7.58 (± 7.17) 6.78 (± 5.32) 10.07 (± 7.54) lessness (SD) m.v. = 1 m.v. = 1 t (sig) 1.924 (0.081) 2.964 (0.009) − 1.334 (0.207) *All significant effects but PCS Magnification in MBSR + TAU group were maintained after Bonferroni correction was applied TAU​, treatment as usual; MBSR, mindfulness-based stress reduction treatment; FibroQoL, psychoeducational programme; VAS, visual analogue scale; FIQR, revised fibromyalgia impact questionnaire; PCS, pain catastrophising scale; m.v., missing value. Alpha = 0.05 positive relationship in the case of the TAU only treatment at baseline was positively correlated with the ACC and the (Fig. 5). That is, large reductions in perceived pain corre- AI across patients. However, this relationship was not main- sponded to small rCBF reductions for both treatment groups, tained following treatment. We observed that MBSR and while large perceived pain reductions were associated with FibroQoL programmes disrupted the positive relationship large rCBF reductions for the control group. between changes in pain scores and changes in rCBF fol- lowing treatment in the insula. Overall, these results provide preliminary evidence of the potential of arterial spin labelled Discussion fMRI as a useful tool to study FMS pathophysiology and mechanisms of action of nonpharmacological therapies. This work aimed to further understand the mechanisms of FibroQoL caused reductions in rCBF in the amygdala, action of two non-pharmacological treatments for FMS unlike MBSR or TAU only. Moreover, changes in amygdala delivered to female FMS patients, as indexed by changes rCBF on the FibroQoL group may relate to a disengage- in rCBF. We identified that the amygdala is differentially ment between negative emotional processing and activity in affected by these therapies, experiencing reductions in rCBF this area, evidenced by the alteration of the linear relation- after 2 months of FibroQoL training programme but not after ship between anxiety and catastrophizing scores and rCBF MBSR training; we also established a relationship between after treatment. In healthy populations, the amygdala is well rCBF in the amygdala and patients’ anxiety and pain cata- known for its role in stress-induced analgesia (Tershner & strophizing within the FibroQoL group. Our second aim was Helmstetter, 2000), pain anticipation and vigilance (Davis to assess whether rCBF is a good marker of individual dif- & Whalen, 2001), and it is a key node involved in pain pro- ferences in FMS symptoms; we found that perceived pain cessing (Veinante et al., 2013). Perfusion in the amygdala 13 456 Mindfulness (2022) 13:449–461 Fig. 3  Summary of results from correlation analyses between rCBF roQoL, psychoeducational programme; HADS, hospital anxiety and in the amygdala bilaterally and behavioural measures for TAU + Fib- depression scale; PCS, pain catastrophising scale; VAS, visual ana- roQoL group. ROI, region of interest. TAU, treatment as usual; Fib- logue scale. Alpha = 0.05 Fig. 4  Results from voxel-wise correlation analysis between whole- wise error correction at alpha = 0.05 post small volume correction in brain baseline perfusion and baseline VAS scores. Initial height the subgenual ACC (top panel) and the anterior insula (bottom panel) threshold = 0.005. Results are significant at peak level after family- 13 Mindfulness (2022) 13:449–461 457 Fig. 5  Comparison of regres- sion slopes between delta VAS scores and delta CBF across treatment groups. Initial height threshold = 0.005. Results are significant at peak level after family-wise error correction at alpha = 0.05 post small volume correction in the anterior insula did not correlate with VAS estimates of pain intensity at any network is exaggerated by high pain catastrophizing (Jiang point. While this may present as a counterintuitive finding, et al., 2016). One may therefore speculate that basal amyg- the complex relationship between the amygdala and clinical dala activity marks the modulation capacity of this connec- pain (Neugebauer, 2007) and depression and anxiety scores tivity pathway due to treatment, ultimately influencing the in FMS has been previously debated (Pujol et al., 2014). A level of ruminating thoughts on pain. morphometric study reported GM reductions in the amyg- We did not obtain similar results within the MSBR group. dala in FMS patients compared to controls that were not Mindfulness practice can improve the processing of affective related to pain or disability scores (Burgmer et al., 2009) stimuli following a short intervention (Allen et al., 2012), suggesting that physiological and emotional features in evidenced in this study by the reductions in pain catastro- FMS depend upon separate, non-linearly related channels. phizing levels in the MBSR group; however, reductions in While we cannot refute the mediating role of the amygdala amygdala reactivity to negative stimuli and pain seem to in pain symptoms via the descending pain limbic pathway only take place in experienced meditators (Kral et al., 2018; to the periaqueductal grey (Linnman et al., 2012), our data Lutz et al., 2013). It is therefore plausible that changes in suggests that its role in modulating clinical improvement the perfusion of the amygdala might start to be visible only concerns mainly automatic anxiety/stress regulation; specu- at later stages of the intervention. Future imaging studies latively, this regulation may occur via a reorganization of might shed further light in this direction. We did not find any the connectivity with other areas, such as the ventromedial relationship between FIQR and rCBF. FIQR is considered prefrontal cortex due to changes in cognitive control (Kral the gold standard measure of functional impairment in FMS; et al., 2018), the AI as one of the salience network path- however, it is worth noting that the FIQR covers a broad ways (Menon & Uddin, 2010), and the hippocampus in the spectrum of FMS features (i.e. function, overall impact and context of forming and retrieving memories and learning symptoms), each one likely specific to different phenotypes (Davis & Whalen, 2001). The extent of amygdala modula- of patients as well as separate brain regions and neurophysi- tion also appears to depend on baseline anxiety levels, and ological processes (Pérez-Aranda, Andrés-Rodríguez, et al., greater amygdala rCBF at baseline correlated with higher 2019). Given this, it is a reasonable outcome that rCBF in PCS improvement. Previous evidence suggests that dysfunc- our three separate ROIs could not capture differences in tional FC between the amygdala and the central executive FIQR after treatment or individual FIQR differences. 13 458 Mindfulness (2022) 13:449–461 Our second objective was to determine whether rCBF stimuli discrimination as well improved affective processing could be a sensible predictor of pain in this set of FMS leads to a stronger engagement of the AI at rest. This is also patients. In line with our hypothesis, we observed a positive evidenced by the fact that although the control group did correlation between VAS pain scores and rCBF in the AI and report reductions in pain symptoms (discussed below), we the sgACC. The AI has been widely known for modulating observed the opposite tendency. In fact, this VAS reduction emotional processing and meaning attribution of pain, and in the control group was a surprising result. Nevertheless, it is involved specifically in integrating emotional, rather placebo effects in fibromyalgia RCTs have been reported than sensory, aspects of pain in FMS (Giesecke et al., 2005). (Chen et al., 2017), and higher levels of pain catastrophising The ACC receives nociceptive input from the amygdala, can enhance this effect (Sullivan et al., 2008). Together with insula and somatosensory cortices (Bliss et al., 2016) and the absence of any significant result at the brain level that preclinical studies have shown that it is an important and relates to treatment efficacy, we argue that this reduction in separate hub for mood disorders associated with chronic pain may be the consequence of placebo responses. Another pain (Barthas et al., 2015). This indicates that ACC activity possibility is that this overall reduction in perceived pain may well be linked to changes in psychological symptoms ratings is a product of regression to the mean phenomenon, in FMS. In fact, the correlation between pain scores and judging by the distribution of baseline VAS scores in the rCBF in these areas disappeared after treatment, suggest- TAU only group (supplementary Fig. 1). ing functional reorganization within the salience and pain networks. In this subsample from the main EUDAIMON trial, there was not a significant reduction in perceived pain Limitations and Future Research Directions after treatment across MBSR and FibroQoL groups. From a therapeutic perspective, this was not the most desirable We encountered two major difficulties in this study. First, outcome; however, it is important to note that the purpose the limited coverage across the brain of the rCBF maps pre- of this substudy was not to assess the efficacy of mind–body vented us from adopting a more exploratory, whole-brain interventions in chronic pain relief, but rather to investigate analysis approach and forced us to discard a significant pro- these techniques at a mechanistic level. Moreover, the VAS portion of the acquired data. Future research may elucidate scores reflected perceived pain on the day of the scan, which whether the distinct brain perfusion patterns across therapies did not necessarily reflect intervention effects on general extend to other brain regions. Specifically, we hypothesize pain experience. Notably, our regression slope compari- that the somatosensory cortex would experience perfusion son analysis showed evidence for a shift in the association reductions after mind–body treatment, more so in the case between changes in pain and changes in rCBF after MBSR of FibroQoL than following MBSR, since FibroQoL focuses and FibroQoL in the AI. In contrast with our initial assump- on body relaxation. Similarly, we would expect perfusion in tions, large reductions in pain corresponded to little rCBF areas focused on top-down pain modulation, such as pain- reductions after treatment. Although reductions in pain in related nuclei of the brainstem (PAG, VTA, red nucleus, FMS after a nonpharmacological intervention have been RVM) to experience relative increases following MBSR linked to decreased functional connectivity of the AI with treatment as a product of improved self-awareness, as well the default mode network (Napadow et al., 2012) and the as attentional and cognitive control. The second limitation somatosensory cortex (Lazaridou et al., 2017), it is impor- (and largely a consequence of the first one) is the sample size tant to note that the role of the AI in acute and chronic pain within each one of our treatment groups. Low sample sizes extends beyond sensory processing; it is a prime node of can affect the reproducibility of results and inflate effect the salience network (Menon & Uddin, 2010) and is key size estimations (Dumas-Mallet et al., 2017). Nevertheless, in the integration of context meaning and emotional states statistics such as Hedge’s G can outperform more ordinary to make decisions on potentially harmful or pain stimuli effect size statistics on those cases where group samples (Wiech et al., 2010). Moreover, these functions can became are unequal and below 20. Hedge’s G values for the present increasingly maladaptive with age in chronic pain patients study show that our results yielded medium to large effect (Ceko et al., 2013). In general, mind–body techniques focus sizes (Table 2), providing some confidence on the present on improving emotion processing and judgment and on results. Future replications of this study on larger samples eliminating attention biases in relation to our own physi- may shed further light on this matter. ological sensations (Wu et al., 2019), all roles which relate to AI activity. We therefore suggest that rather than ‘reduc- Supplementary Information The online version contains supplemen- tary material available at https://d​ oi.o​ rg/1​ 0.1​ 007/s​ 12671-0​ 21-0​ 1806-2. ing hyperactivity’ of the AI, FibroQoL and MBSR treat- ments provoke a retraining on this area, and while aberrant Acknowledgements We would like to express our sincere thanks to connectivity with other areas involved in clinical pain may Dr Tom Barrick at St George’s Hospital Medical School for helping be diminished, greater control over internal and external us process the Philips ASL data. 13 Mindfulness (2022) 13:449–461 459 Table 2  Summary of effect size calculations Contrast N Statistic Significance Cohen’s d Hedge’s G Effect Paired t-test (baseline vs follow-up) for TAU + FibroQoL group 18 T = 3.74 pFWE = 0.022 0.632 0.625 Medium Correlation between HADS baseline and amygdala baseline CBF 18 r =  − 0.57 p = 0.028 1.387 1.372 Large–very large Correlation between PCS baseline and amygdala baseline CBF 18 r = 0.56 p = 0.016 1.351 1.337 Large–very large FWE, family-wise error Author Contribution SM: contributed to the formal analysis, methodol- References ogy, software, visualization, and writing of the original draft. OGOD: contributed to the formal analysis, software, validation, and review and Allen, M., Dietz, M., Blair, K. S., van Beek, M., Rees, G., Vestergaard- editing of the manuscript. MAH: contributed to the conceptualization, Poulsen, P., Lutz, A., & Roepstorff, A. (2012). Cognitive-Affective formal analysis, methodology, supervision, validation, and review and Neural Plasticity following Active-Controlled Mindfulness Inter- editing of the manuscript. AFS: contributed to the conceptualization, vention. Journal of Neuroscience, 32(44), 15601–15610. methodology, data curation, funding acquisition, investigation, project Aman, M. M., Yong, R. J., Kaye, A. D., & Urman, R. D. (2018). Evi- administration, supervision, validation, and review and editing of the dence-based non-pharmacological therapies for fibromyalgia. manuscript. JVL: contributed to the conceptualization, methodology, Current Pain and Headache Reports, 22(5), 1–5. data curation, funding acquisition, investigation, project administra- Andrés-Rodríguez, L., Borràs, X., Feliu-Soler, A., Pérez-Aranda, tion, supervision, validation, and review and editing of the manuscript. A., Rozadilla-Sacanell, A., Montero-Marin, J., et al. (2019). MAH, AFS and JVL share senior authorship. Immune-inflammatory pathways and clinical changes in fibromy- algia patients treated with Mindfulness-Based Stress Reduction Funding SM and MAH are funded by a Medical Research Council (MBSR): A randomized, controlled clinical trial. Brain, Behavior, Experimental Medicine Challenge Grant (MR/N026969/1). SM, MAH and Immunity, 80, 109–119. and OOD are also supported by the NIHR Biomedical Research Centre Barthas, F., Sellmeijer, J., Hugel, S., Waltisperger, E., Barrot, M., & for Mental Health at the South London and Maudsley NHS Trust. AFS Yalcin, I. (2015). The Anterior Cingulate Cortex Is a Critical acknowledges the funding from the Serra Húnter program (Generalitat Hub for Pain-Induced Depression. Biological Psychiatry, 77(3), de Catalunya; reference number UAB-LE-8015). JVL has a “Miguel 236–245. Servet” research contract from the ISCIII (CPII19/00003). Bennett, R., & Nelson, D. (2006). Cognitive behavioural therapy medical research council,MR/N026969/1,Matthew Alex- for fibromyalgia. Nature Clinical Practice Rheumatology, 2(8), ander Howard,serra húnter program,UAB-LE-8015,Albert 416–424. Feliu-Soler,“miguel ser vet” research contract from the Bennett, R. M., Friend, R., Jones, K. D., Ward, R., Han, B. K., & isciii,CPII19/00003,Juan Vicente Luciano Ross, R. L. (2009). The revised fibromyalgia impact question- naire (FIQR): Validation and psychometric properties. Arthritis Declarations Research & Therapy, 11(4), 1–14. Bliss, T. V., Collingridge, G. L., Kaang, B. K., & Zhuo, M. (2016). Synaptic plasticity in the anterior cingulate cortex in acute and Data Accessibility Statement The data that support the findings of this chronic pain. Nature Reviews Neuroscience, 17(8), 485–496. study are available on request from the corresponding author. The data Burgmer, M., Gaubitz, M., Konrad, C., Wrenger, M., Hilgart, S., Heuft, are not publicly available due to privacy or ethical restrictions. G., & Pfleiderer, B. (2009). Decreased grey matter volumes in the The study was performed in accordance with the ethical standards cingulo-frontal cortex and the amygdala in patients with fibromy- specified in the 1964 Declaration of Helsinki and its subsequent up- algia. Psychosomatic Medicine, 71(5), 566–573. dates. The FSJD Research Committee Board evaluated and approved Ceko, M., Bushnell, M. C., Fitzcharles, M. A., & Schweinhardt, P. the study protocol in May 2015 (PIC-102–15). (2013). Fibromyalgia interacts with age to change the brain. Neu- roImage Clinical, 3, 249–260. Consent to Participate Informed consent was obtained from all indi- Chen, X., Zou, K., Abdullah, N., Whiteside, N., Sarmanova, A., vidual participants included in the study. Doherty, M., & Zhang, W. (2017). The placebo effect and its determinants in fibromyalgia: meta-analysis of randomised con- trolled trials. Clinical Rheumatology, 36(7), 1623–1630. Conflict of Interest The authors declare no competing interests. Cummiford, C. M., Nascimento, T. D., Foerster, B. R., Clauw, D. J., Zubieta, J. K., Harris, R. E., & DaSilva, A. F. (2016). Changes in Open Access This article is licensed under a Creative Commons Attri- resting state functional connectivity after repetitive transcranial bution 4.0 International License, which permits use, sharing, adapta- direct current stimulation applied to motor cortex in fibromyalgia tion, distribution and reproduction in any medium or format, as long patients. Arthritis Research & Therapy, 18(1), 1–12. as you give appropriate credit to the original author(s) and the source, Davis, M., & Whalen, P. J. (2001). The amygdala: Vigilance and emo- provide a link to the Creative Commons licence, and indicate if changes tion. Molecular Psychiatry, 6(1), 13–34. were made. The images or other third party material in this article are Derbyshire, S. W., Whalley, M. G., & Oakley, D. A. (2009). Fibro- included in the article's Creative Commons licence, unless indicated myalgia pain and its modulation by hypnotic and non-hypnotic otherwise in a credit line to the material. If material is not included in suggestion: An fMRI analysis. European Journal of Pain, 13(5), the article's Creative Commons licence and your intended use is not 542–550. permitted by statutory regulation or exceeds the permitted use, you will Dumas-Mallet, E., Button, K. S., Boraud, T., Gonon, F., & Munafò, M. need to obtain permission directly from the copyright holder. To view a R. (2017). Low statistical power in biomedical science: a review copy of this licence, visit http://​creat​iveco​mmons.​org/​licen​ses/​by/4.​0/. 13 460 Mindfulness (2022) 13:449–461 of three human research domains. Royal Society Open Science, Kabat-Zinn, J. (1982). An outpatient program in behavioral medicine 4(2), 160254. for chronic pain patients based on the practice of mindfulness Fallon, N., Chiu, Y., Nurmikko, T., & Stancak, A. (2016). Functional meditation: Theoretical considerations and preliminary results. connectivity with the default mode network is altered in fibromy- General Hospital Psychiatry, 4(1), 33–47. algia patients. PLOS ONE, 11(7), e0159198. Kim, J., Loggia, M. L., Cahalan, C. M., Harris, R. E., Beissner, F., Gar- Fan, L., Li, H., Zhuo, J., Zhang, Y., Wang, J., Chen, L., Yang, Z., cia, R. G., Kim, H., Barbieri, R., Wasan, A. D., Edwards, R. R., & Chu, C., Xie, S., Laird, A. R., Fox, P. T., Eickhoff, S. B., Yu, Napadow, V. (2015). The Somatosensory Link in Fibromyalgia: C., & Jiang, T. (2016). The Human Brainnetome Atlas: A New Functional Connectivity of the Primary Somatosensory Cortex Is Brain Atlas Based on Connectional Architecture. Cerebral Cortex, Altered by Sustained Pain and Is Associated With Clinical/Auto- 26(8), 3508–3526. nomic Dysfunction. Arthritis & Rheumatology, 67(5), 1395–1405. Feliu-Soler, A., Borràs, X., Peñarrubia-María, M. T., Rozadilla- Kong, J., Wolcott, E., Wang, Z., Jorgenson, K., Harvey, W. F., Tao, J., Sacanell, A., D’Amico, F., Moss-Morris, R., Howard, M. A., Rones, R., & Wang, C. (2018). Altered resting state functional Fayed, N., Soriano-Mas, C., Puebla-Guedea, M., Serrano-Blanco, connectivity of the cognitive control network in fibromyalgia and A., Pérez-Aranda, A., Tuccillo, R., & Luciano, J. V. (2016). Cost- the modulation effect of mind-body intervention. Brain Imaging utility and biological underpinnings of Mindfulness-Based Stress and Behavior, 13(2), 482–492. Reduction (MBSR) versus a psychoeducational programme (Fib- Kral, T. R., Schuyler, B. S., Mumford, J. A., Rosenkranz, M. A., Lutz, roQoL) for fibromyalgia: a 12-month randomized controlled trial A., & Davidson, R. J. (2018). Impact of short-and long-term (EUDAIMON study). BMC Complementary and Alternative mindfulness meditation training on amygdala reactivity to emo- Medicine, 16(1). tional stimuli. NeuroImage, 181, 301–313. García-Campayo, J., Rodero, B., del Hoyo, Y. L., Luciano, J. V., Alda, Langley, G. B., & Sheppeard, H. (1985). The visual analogue scale: M., & Gili, M. (2010). Validation of a Spanish language version of Its use in pain measurement. Rheumatology International, 5(4), the pain self-perception scale in patients with fibromyalgia. BMC 145–148. Musculoskeletal Disorders, 11(1), 1–7. Lazaridou, A., Kim, J., Cahalan, C. M., Loggia, M. L., Franceschelli, Garcia-Palacios, A., Herrero, R., Vizcaíno, Y., Belmonte, M. A., O., Berna, C., Schur, P., Napadow, V., & Edwards, R. R. (2017). Castilla, D., Molinari, G., Baños, R. M., & Botella, C. (2015). Effects of cognitive-behavioral therapy (CBT) on brain connec- Integrating Virtual Reality With Activity Management for the tivity supporting catastrophizing in fibromyalgia. The Clinical Treatment of Fibromyalgia. The Clinical Journal of Pain, 31(6), Journal of Pain, 33(3), 215–221. 564–572. Linnman, C., Moulton, E. A., Barmettler, G., Becerra, L., & Borsook, Giesecke, T., Gracely, R. H., Williams, D. A., Geisser, M. E., Petzke, D. (2012). Neuroimaging of the periaqueductal gray: State of the F. W., & Clauw, D. J. (2005). The relationship between depres- field. NeuroImage, 60(1), 505–522. sion, clinical pain, and experimental pain in a chronic pain cohort. Loggia, M. L., Segerdahl, A. R., Howard, M. A., & Tracey, I. (2019). Arthritis & Rheumatism, 52(5), 1577–1584. Imaging clinically relevant pain states using arterial spin labeling. Goldenberg, D. L. (2009). Diagnosis and differential diagnosis of fibro- Pain reports, 4(4), e750. myalgia. The American Journal of Medicine, 122(12), S14–S21. Luciano, J. V., Aguado, J., Serrano-Blanco, A., Calandre, E. P., & Häuser, W., Ablin, J., Fitzcharles, M. A., Littlejohn, G., Luciano, J. Rodriguez-Lopez, C. M. (2013a). Dimensionality, reliability, and V., Usui, C., & Walitt, B. (2015). Fibromyalgia. Naturerreviews validity of the revised fibromyalgia impact questionnaire in two Disease Primers, 1(1), 1–16. Spanish samples. Arthritis Care & Research, 65(10), 1682–1689. Hedges, L. V., & Olkin, I. (2014). Statistical methods for meta-analysis. Luciano, J. V., Sabes-Figuera, R., Cardeñosa, E., Peñarrubia-María, T., Academic press. & M., Fernández-Vergel, R., García-Campayo, J., Knapp, M., & Hsiao, F. J., Chen, W. T., Ko, Y. C., Liu, H. Y., Wang, Y. F., Chen, S. Serrano-Blanco, A. (2013b). Cost-Utility of a Psychoeducational P., & Wang, S. J. (2020). Neuromagnetic Amygdala Response to Intervention in Fibromyalgia Patients Compared With Usual Care. Pain-Related Fear as a Brain Signature of Fibromyalgia. Pain and The Clinical Journal of Pain, 29(8), 702–711. Therapy, 9(2), 765–781. Luciano, J. V., Barrada, J. R., Aguado, J., Osma, J., & García-Campayo, Jensen, K. B., Kosek, E., Petzke, F., Carville, S., Fransson, P., Mar- J. (2014). Bifactor analysis and construct validity of the HADS: cus, H., Williams, S. C. R., Choy, E., Giesecke, T., Mainguy, Y., A cross-sectional and longitudinal study in fibromyalgia patients. Gracely, R., & Ingvar, M. (2009). Evidence of dysfunctional pain Psychological Assessment, 26(2), 395. inhibition in Fibromyalgia reflected in rACC during provoked Lutz, A., McFarlin, D. R., Perlman, D. M., Salomons, T. V., & David- pain. Pain, 144(1), 95–100. son, R. J. (2013). Altered anterior insula activation during antici- Jensen, K. B., Berna, C., Loggia, M. L., Wasan, A. D., Edwards, R. R., pation and experience of painful stimuli in expert meditators. & Gollub, R. L. (2012a). The use of functional neuroimaging to NeuroImage, 64, 538–546. evaluate psychological and other non-pharmacological treatments Martin-Ausero, A. (2008). Con rumbo propio: responder a situaciones for clinical pain. Neuroscience Letters, 520(2), 156–164. de crisis. Plataforma Editorial. Jensen, K. B., Kosek, E., Wicksell, R., Kemani, M., Olsson, G., Merle, Menon, V., & Uddin, L. Q. (2010). Saliency, switching, attention and J. V., Kadetoff, D., & Ingvar, M. (2012b). Cognitive Behavioral control: A network model of insula function. Brain Structure and Therapy increases pain-evoked activation of the prefrontal cortex Function, 214(5–6), 655–667. in patients with fibromyalgeia. Pain, 153(7), 1495–1503. Moana-Filho, E. J., Tchivileva, I. E., & Gracely, R. (2013). Assessment Jensen, K. B., Loitoile, R., Kosek, E., Petzke, F., Carville, S., Fransson, of resting-state cerebral blood flow in fibromyalgia patients using P., Marcus, H., Williams, S. C., Choy, E., Mainguy, Y., Vitton, arterial spin labelling.: 43rd Annual Meeting of the Society for O., Gracely, R. H., Gollub, R., Ingvar, M., & Kong, J. (2012c). Neuroscience. Experts.umn.edu. Patients with fibromyalgia display less functional connectivity in Moher, D., Hopewell, S., Schulz, K. F., Montori, V., Gøtzsche, P. the brain's pain inhibitory network. Molecular Pain, 8, 32. C., Devereaux, P. J., Elbourne, D., Egger, M., & Altman, D. G. Jiang, Y., Oathes, D., Hush, J., Darnall, B., Charvat, M., Mackey, S., & (2012). CONSORT 2010 explanation and elaboration: Updated Etkin, A. (2016). Perturbed connectivity of the amygdala and its guidelines for reporting parallel group randomized trials. Inter- subregions with the central executive and default mode networks national Journal of Surgery, 10(1), 28–55. in chronic pain. Pain, 157(9), 1970. Müller, M., Wüthrich, F., Federspiel, A., Wiest, R., Egloff, N., Reichen- bach, S., Exadaktylos, A., Jüni, P., Curatolo, M., & Walther, S. 13 Mindfulness (2022) 13:449–461 461 (2021). Altered central pain processing in fibromyalgia—A mul- Schulz, K. F., Altman, D. G., Moher, D., & CONSORT Group. (2010). timodal neuroimaging case-control study using arterial spin label- CONSORT 2010 statement: updatedguidelines for reporting par- ling. PLOS ONE, 16(2), e0235879. allel group randomised trials. BMJ (Clinical Research ed.), 340. Napadow, V., Kim, J., Clauw, D. J., & Harris, R. E. (2012). Brief Sprenger, C., Bingel, U., & Büchel, C. (2011). Treating pain with pain: report: Decreased intrinsic brain connectivity is associated with Supraspinal mechanisms of endogenous analgesia elicited by het- reduced clinical pain in fibromyalgia. Arthritis & Rheumatism, erotopic noxious conditioning stimulation. Pain, 152(2), 428–439. 64(7), 2398–2403. Stucki, M. O. G. (2000). Physical therapy in the treatment of fibromy- Neugebauer, V. (2007). The amygdala: Different pains, different mech- algia. Scandinavian Journal of Rheumatology, 29(113), 78–85. anisms. Pain, 127(1–2), 1. Sullivan, M. J., Bishop, S. R., & Pivik, J. (1995). The pain catastrophiz- Neugebauer, V., Li, W., Bird, G. C., & Han, J. S. (2004). The amygdala ing scale: Development and validation. Psychological Assessment, and persistent pain. The Neuroscientist, 10(3), 221–234. 7(4), 524. Nüesch, E., Häuser, W., Bernardy, K., Barth, J., & Jüni, P. (2013). Sullivan, M. J., Lynch, M. E., Clark, A. J., Mankovsky, T., & Sawynok, Comparative efficacy of pharmacological and non-pharmaco- J. (2008). Catastrophizing and treatment outcome: Differential logical interventions in fibromyalgia syndrome: Network meta- impact on response to placebo and active treatment outcome. Con- analysis. Annals of the Rheumatic Diseases, 72(6), 955–962. temporary Hypnosis, 25(3–4), 129–140. Pérez-Aranda, A., Andrés-Rodríguez, L., Feliu-Soler, A., Núñez, C., Sumpton, J. E., & Moulin, D. E. (2008). Fibromyalgia: Presentation Stephan-Otto, C., Pastor-Mira, M. A., López-Roig, S., Peña- and management with a focus on pharmacological treatment. Pain coba, C., Calandre, E. P., Slim, M., Salgueiro, M., Feixas, G., Research and Management, 13(6), 477–483. & Luciano, J. V. (2019a). Clustering a large Spanish sample of Tershner, S. A., & Helmstetter, F. J. (2000). Antinociception produced patients with fibromyalgia using the Fibromyalgia Impact Ques- by mu opioid receptor activation in the amygdala is partly depend- tionnaire-Revised: Differences in clinical outcomes, economic ent on activation of mu opioid and neurotensin receptors in the costs, inflammatory markers, and gray matter volumes. Pain, ventral periaqueductal gray. Brain Research, 865(1), 17–26. 160(4), 908–921. Veinante, P., Yalcin, I., & Barrot, M. (2013). The amygdala between Pérez-Aranda, A., D’Amico, F., Feliu-Soler, A., McCracken, L. M., sensation and affect: A role in pain. Journal of Molecular Psy- Peñarrubia-María, M. T., Andrés-Rodríguez, L., Angarita-Osorio, chiatry, 1(1), 1–14. N., Knapp, M., García-Campayo, J., & Luciano, J. V. (2019b). Wiech, K., Lin, C. S., Brodersen, K. H., Bingel, U., Ploner, M., & Cost-Utility of Mindfulness-Based Stress Reduction for Fibro- Tracey, I. (2010). Anterior insula integrates information about myalgia versus a Multicomponent Intervention and Usual Care: salience into perceptual decisions about pain. Journal of Neurosci- A 12-Month Randomized Controlled Trial (EUDAIMON Study). ence, 30(48), 16324–16331. Journal of Clinical Medicine, 8(7), 1068. Wolfe, F., Schmukler, J., Jamal, S., Castrejon, I., Gibson, K. A., Srini- Pérez-Aranda, A., Feliu-Soler, A., Montero-Marín, J., García-Cam- vasan, S., Häuser, W., & Pincus, T. (2019). Diagnosis of fibromy- payo, J., Andrés-Rodríguez, L., Borràs, X., Rozadilla-Sacanell, algia: Disagreement between fibromyalgia criteria and clinician- A., Peñarrubia-Maria, M. T., Angarita-Osorio, N., McCracken, based fibromyalgia diagnosis in a University Clinic. Arthritis Care L. M., & Luciano, J. V. (2019c). A randomized controlled effi- & Research, 71(3), 343–351. cacy trial of mindfulness-based stress reduction compared with an Wu, R., Liu, L. L., Zhu, H., Su, W. J., Cao, Z. Y., Zhong, S. Y., Liu, X. active control group and usual care for fibromyalgia: The EUDAI- H., & Jiang, C. L. (2019). Brief mindfulness meditation improves MON study. Pain, 160(11), 2508–2523. emotion processing. Frontiers in Neuroscience, 13, 1074. Picard, P., Jusseaume, C., Boutet, M., Dualé, C., Mulliez, A., & Aublet- Young, K. S., van der Velden, A. M., Craske, M. G., Pallesen, K. J., Cuvellier, B. (2013). Hypnosis for management of fibromyalgia. Fjorback, L., Roepstorff, A., & Parsons, C. E. (2018). The impact International Journal of Clinical and Experimental Hypnosis, of mindfulness-based interventions on brain activity: A systematic 61(1), 111–123. review of functional magnetic resonance imaging studies. Neuro- Pujol, J., Macià, D., Garcia-Fontanals, A., Blanco-Hinojo, L., López- science & Biobehavioral Reviews, 84, 424–433. Solà, M., Garcia-Blanco, S., Poca-Dias, V., Harrison, B. J., Con- Zeidan, F., Salomons, T., Farris, S. R., Emerson, N. M., Adler-Neal, treras-Rodríguez, O., Monfort, J., Garcia-Fructuoso, F., & Deus, A., Jung, Y., & Coghill, R. C. (2018). Neural mechanisms sup- J. (2014). The contribution of sensory system functional con- porting the relationship between dispositional mindfulness and nectivity reduction to clinical pain in fibromyalgia. Pain, 155(8), pain. Pain, 159(12), 2477. 1492–1503. Zigmond, A. S., & Snaith, R. P. (1983). The hospital anxiety and Schrepf, A., Harper, D. E., Harte, S. E., Wang, H., Ichesco, E., Hamp- depression scale. Acta Psychiatrica Scandinavica, 67(6), 361–370. son, J. P., Zubieta, J.-K., Clauw, D. J., & Harris, R. E. (2016). Endogenous opioidergic dysregulation of pain in fibromyalgia: A Publisher's Note Springer Nature remains neutral with regard to PET and fMRI study. Pain, 157(10), 2217–2225. jurisdictional claims in published maps and institutional affiliations. 13