Content uploaded by Albert Feliu-Soler
Author content
All content in this area was uploaded by Albert Feliu-Soler on Jan 09, 2025
Content may be subject to copyright.
Content uploaded by Albert Feliu-Soler
Author content
All content in this area was uploaded by Albert Feliu-Soler on Jan 04, 2025
Content may be subject to copyright.
Full length article
Immune-inammatory effects of the multicomponent intervention
FIBROWALK in outdoor and online formats for patients with bromyalgia
S`
onia Ferr´
es
a,b,1
, Mayte Serrat
c,1,*
, William Auer
b
, Estíbaliz Royuela-Colomer
d
,
Míriam Almirall
c
, Andrea Lizama-Lefno
e
, Jo Nijs
f,g,h
, Michael Maes
i
, Juan V. Luciano
d,j,k,2
,
Xavier Borr`
as
b,k,2
, Albert Feliu-Soler
d,k,2,*
a
Escoles Universit`
aries Gimbernat, Autonomous University of Barcelona, Bellaterra, Spain
b
Department of Basic, Developmental and Educational Psychology, Faculty of Psychology, Autonomous University of Barcelona, Bellaterra, Spain
c
Unitat d’Expertesa en Síndromes de Sensibilitzaci´
o Central, Servei de Reumatologia, Vall d’Hebron Hospital Universitari, Vall d’Hebron Barcelona Hospital Campus,
Barcelona, Spain
d
Department of Clinical and Health Psychology, Faculty of Psychology, Autonomous University of Barcelona, Bellaterra, Spain
e
Department of Development and Postgraduate, Autonomous University of Chile, Chile
f
Pain in Motion Research Group (PAIN), Department of Physiotherapy, Human Physiology and Anatomy, Faculty of Physical Education &Physiotherapy, Vrije
Universiteit Brussel, Brussels, Belgium
g
Department of Health and Rehabilitation, Unit of Physiotherapy, Institute of Neuroscience and Physiology, Sahlgrenska Academy, University of Gothenburg, Sweden
h
Chronic Pain Rehabilitation, Department of Physical Medicine and Physiotherapy, University Hospital Brussels, Belgium
i
Sichuan Provincial Center for Mental Health, University of Electronic Science and Technology of China, Chengdu, China
j
Teaching, Research &Innovation Unit, Parc Sanitari Sant Joan de D´
eu, St. Boi de Llobregat, Spain
k
Centre for Biomedical Research in Epidemiology and Public Health (CIBERESP), Madrid, Spain
ARTICLE INFO
Keywords:
Fibromyalgia
Multicomponent Intervention
FIBROWALK
Online
Outdoor
Cytokines
Brain-derived neurotrophic factor
ABSTRACT
The multicomponent intervention FIBROWALK integrates pain science education (PSE), therapeutic exercise,
cognitive behavioral therapy (CBT), and mindfulness training for treating bromyalgia (FM). This study inves-
tigated the effects of the FIBROWALK in online (FIBRO-On) and outdoor (FIBRO-Out) formats compared to
treatment-as-usual (TAU) on core clinical variables along with serum immune-inammatory biomarkers and
brain-derived neurotrophic factor (BDNF). Furthermore, the predictive value of these biomarkers on clinical
response to FIBROWALK was also evaluated. 120 participants were randomly divided into three groups: TAU,
TAU +FIBRO-On or TAU +FIBRO-Out. Clinical and blood assessments were conducted pre-post treatment. Both
FIBRO-Out and FIBRO-On showed effectiveness (vs TAU) by improving functional impairment and kinesi-
ophobia. Individuals allocated to FIBRO-Out (vs TAU) additionally showed decreases in pain, fatigue, depressive
symptoms, and serum IL-6 and IL-10 levels along with IL-6/IL-4 ratio; patients allocated to FIBRO-On only
showed a less stepped increase in IL-6 compared to TAU. An exaggerated pro-inammatory prole along with
higher levels of BDNF at baseline predicted greater clinical improvements in both active treatment arms. Our
results suggest that FIBROWALK −in online and outdoor formats- is effective in individuals with FM and has
signicant immune regulatory effects in FM patients, while immune-inammatory pathways and BDNF levels
may in part predict its clinical effectiveness.
Trial registration number NCT05377567 (clinicaltrials.gov).
1. Introduction
Fibromyalgia (FM) stands out as a highly prevalent syndrome,
affecting approximately 2.7 % of the global population, with a
particularly notable prevalence among women aged 40–50 years
(Heidari et al. 2017; H¨
auser et al. 2015). The hallmark features of FM
include persistent musculoskeletal pain, fatigue, and sleep disturbances,
frequently intertwined with anxiety and depressive disorders
* Corresponding authors.
E-mail addresses: mayte.serrat@vallhebron.cat (M. Serrat), albert.feliu@uab.cat (A. Feliu-Soler).
1
These authors contributed equally and should be considered as co-rst authors.
2
These authors share senior authorship.
Contents lists available at ScienceDirect
Brain Behavior and Immunity
journal homepage: www.elsevier.com/locate/ybrbi
https://doi.org/10.1016/j.bbi.2024.12.149
Received 11 June 2024; Received in revised form 10 December 2024; Accepted 21 December 2024
Brain, Behavior, and Immunity 125 (2025) 184–197
Available online 30 December 2024
0889-1591/© 2024 The Author(s). Published by Elsevier Inc. This is an open access article under the CC BY license (
http://creativecommons.org/licenses/by/4.0/ ).
(Lichtenstein et al. 2018). Multiple physiological factors are posited to
underlie the symptomatology of FM, with the central nervous system
assuming a pivotal role, contributing signicantly to the manifestation
and persistence of symptoms (de Tommaso et al. 2022; Sawaddiruk et al.
2017; Sluka and Clauw, 2016). Several studies have reported elevated
levels of pro-inammatory cytokines and chemokines (e.g., IL-1, IL-6,
CXCL8, TNF-
α
) alongside reduced levels of anti-inammatory cytokines
(e.g., IL-4, IL-5, IL-10, IL-13) in individuals with FM (B¨
ackryd et al.,
2017; Rodriguez-Pinto et al., 2014; Sluka and Clauw, 2016). These
ndings support the hypothesis of chronic low-grade systemic immune-
inammatory activation in FM, potentially lowering pain thresholds and
contributing to peripheral nerve sensitization (Rodriguez-Pinto et al.,
2014). Additionally, a systematic review and meta-analysis by
Andr´
es-Rodríguez et al. (2020) highlighted mild immune alterations in
individuals with FM, suggesting an imbalance between upregulated
immune-inammatory and immunoregulatory pathways, marked by
elevated levels of IL-6, IL-17A, and IL-4. Both systemic and neuro-
inammation likely play a role in exacerbating FM symptoms, such as
pain sensitivity, fatigue, and sleep disturbances (Andr´
es-Rodríguez
et al., 2020). Notably, these immune aberrations are linked to dysre-
gulation in biomarkers related to neuronal plasticity, including
brain-derived neurotrophic factor (BDNF) (Deitos et al., 2015). BDNF, a
neurotrophin crucial for neuroplasticity, is involved in pain modulation,
nociception, and hyperalgesia, all of which are disrupted in FM (Xiong
et al., 2024; Nugraha et al., 2012). Elevated BDNF levels in FM patients
contribute to widespread hyperalgesia (Polli et al., 2020), and modu-
lating BDNF through therapeutic interventions is hypothesized to reduce
pain intensity (Di-Bonaventura et al., 2023).
Currently available treatments for FM are not curative and have
limited efcacy (H¨
auser et al., 2015). Concerning pharmacological in-
terventions, the European League Against Rheumatism (EULAR) rec-
ommends using pharmaceuticals primarily for cases involving severe
pain and sleep disturbances (H¨
auser et al. 2015; Macfarlane et al. 2017).
Non-pharmacological strategies demonstrated more ubiquitous effects
than pharmacological ones, often exhibiting slightly larger effect sizes
compared to pharmaceutical options (Perrot and Russell, 2014; Nüesch
et al. 2013; Hong-Baik et al. 2023). These non-pharmacological ap-
proaches encompass diverse interventions such as pain science educa-
tion (PSE), therapeutic physical exercise, cognitive behavioral therapy,
and mindfulness training where their main goal is to alleviate symptoms
and enhance the overall quality of life for individuals living with FM
(Macfarlane et al. 2017). The effectiveness of multicomponent in-
terventions combining exercise, psychological support, and education in
FM treatment is well-supported (Thieme et al. 2017; Sharpe et al. 2020).
These comprehensive approaches are often regarded as the optimal
standard for managing the condition (De Miquel et al. 2010; H¨
auser
et al. 2009; Macfarlane et al. 2017; Rivera et al. 2006; Thieme et al.
2017).
Non-pharmacological interventions, such as CBT, have been shown
to modulate immune responses and reduce inammation, further
highlighting their clinical relevance (Andr´
es-Rodríguez et al. 2019;
Sanada et al., 2015). A systematic review and meta-analysis by Shields
et al. (2020) reported improvements in immune function following
psychosocial interventions, with CBT and multicomponent programs
showing the most signicant effects on reducing pro-inammatory
biomarkers. Other studies suggest that multicomponent interventions,
physical exercise, and dietary modications may have anti-
inammatory effects, particularly on markers like IL-6 and CXCL8
(Sanada et al., 2015). Randomized controlled trials also indicate that
mindfulness and compassion-based interventions can reduce FM symp-
toms and lower serum BDNF levels compared to active control groups
(Montero-Marin et al. 2019; Sanabria-Mazo et al., 2020).
FIBROWALK is a 3-month multicomponent intervention for in-
dividuals with FM combining PSE, therapeutic physical exercise, CBT
and mindfulness added to TAU, which has shown substantial improve-
ments in functionality, pain, kinesiophobia, physical function, fatigue,
anxiety, and depressive symptomatology in FM patients (Serrat et al.
2020, 2021a, 2021b, 2022a). It has demonstrated short-term clinical
effectiveness compared to treatment-as-usual (TAU) in various settings,
including hospitals, outdoor environments, and online platforms,
although the studies conducted until now showed some methodological
weaknesses (Serrat et al. 2022b). Teletherapy approaches like online
FIBROWALK (FIBRO-On) are particularly promising in overcoming
logistical and health barriers, potentially enhancing treatment adher-
ence (Li et al. 2020; Schwamm et al. 2020; White et al. 2022) and were
particularly useful in pandemic times (Serrat et al. 2021a). Similarly,
nature-based therapeutic approaches, exemplied by outdoor FIBRO-
WALK (FIBRO-Out), have been shown to be an effective approach in
improving mental health across various clinical conditions, including
chronic pain and FM (L´
opez-Pousa et al. 2015; Serrat et al. 2020;
Stanhope et al. 2020). In this regard, compared to therapeutic exercise
performed indoors, exercise in natural settings has been associated with
greater feelings of revitalization and positive engagement, reductions in
tension, confusion, anger and depressive symptomatology, and
increased energy (Thompson Coon et al. 2011). In addition, there is
evidence that exercising outdoors (vs. indoors) may also promote
directed attention and social interactions, which may positively inu-
ence future intention to maintain an exercise routine (Rogerson et al.
2016). Furthermore, exposure to nature has been shown to positively
affect immunological health, including reduced expression of
pro-inammatory cytokines and increased levels of anti-inammatory
cytokines (for a review, see Andersen et al., 2021). This suggests that
adapting non-pharmacological interventions to outdoor settings may
offer additional benets in promoting immune-inammatory normali-
zation in patients with FM.
This trial, embedded within the On&Out study (Serrat et al. 2022b),
aimed to assess the short-term effectiveness of adding the multicompo-
nent FIBROWALK intervention, delivered in both outdoor (i.e., FIBRO-
Out) and online (i.e. FIBRO-On) formats, to TAU compared to TAU
alone, and to evaluate their impact on specic serum immune-
inammatory biomarkers and BDNF levels, for which there is some
evidence of alteration in FM. We hypothesized that, compared to TAU,
the interventions would reduce pro-inammatory markers (IL-6, CXCL8,
IL-17A, and hs-CRP) and BDNF levels, while increasing anti-
inammatory cytokines (IL-4 and IL-10), with the outdoor interven-
tion expected to be more effective at reducing pro-inammatory status
and promoting anti-inammatory effects. This exploratory study com-
pares the impact of the same multicomponent program delivered in
different formats, which have not been previously assessed for their
effects on either biomarkers or clinical outcomes in FM. Finally, we
explored the predictive capacity of baseline biomarkers to assess clinical
improvement, aiming to identify potential biomarker proles that could
help predict therapeutic effectiveness and take a step toward more
personalized treatment approaches.
2. Material and methods
2.1. Study design
The current study adopts a pre-post parallel-group, single-blinded
randomized design, employing a computer-generated randomization
list featuring three arms: 1) TAU, 2) TAU plus FIBRO-On, and 3) TAU
plus FIBRO-Out. This study is embedded within a large randomized
controlled trial (RCT) involving a total of 225 participants distributed
across the three study arms described above, with a subsequent 6-month
follow-up (Serrat et al. 2022b), received approval from the Ethics
Committee of Clinical Investigation of the Hospital Universitari Vall
d’Hebron (HUVH) in Barcelona, and was registered at ClinicalTrials.gov
(NCT05377567). The research adhered to ethical standards outlined in
the Declaration of Helsinki (current version: Fortaleza, Brazil, October
2013) and was conducted in accordance with the prescribed protocol
and relevant legal requisites, including Law 14/2007, July 3, 2007, on
S. Ferr´
es et al. Brain Behavior and Immunity 125 (2025) 184–197
185
Biomedical Research.
2.2. Participants
A total of 120 individuals recruited from the HUVH Specialized Unit
for Central Sensitivity Syndromes participated in the study. Previous
RCTs of the FIBROWALK interventions demonstrated effect sizes
ranging from d =0.80 (FIBRO-On; Serrat et al. 2022b) to d =1.83
(FIBRO-Out; Serrat et al. 2020). Based on these results, a minimum of 26
participants per condition (78 in total) with an alpha of 0.05 and a
power of 0.80 was deemed necessary. Considering a 25 % dropout rate,
the minimum nal sample size was estimated at 33 participants per arm.
However, this minimum was slightly increased to 40 per arm after
reviewing previous RCTs on the effects of non-pharmacological treat-
ments in FM on similar immune-inammatory biomarkers (Andr´
es-
Rodríguez et al. 2019; Montero-Marin et al. 2019).
All participants had to meet specic inclusion criteria: 1) adult fe-
males (≥18 years old), 2) diagnosis of FM according to American College
of Rheumatology (ACR) 2010/2011 criteria, and 3) uent in written and
spoken Spanish. General exclusion criteria included: 1) receipt of psy-
chological treatment within the last year or ongoing, 2) comorbid
presence of severe mental disorders (e.g., schizophrenia) or other ter-
minal clinical conditions or scheduled treatments that could disrupt
study follow-up, 3) inability to consistently complete the weekly ses-
sions/modules of the program, and 4) usual contraindications for
measuring immune-inammatory markers in blood (e.g., autoimmune
diseases, recent physical trauma, cold/infection on the day of blood
collection, needle phobia, pregnant or breastfeeding, and using oral or
local corticosteroids, anti-cytokine biologic drugs, or oral contracep-
tives). Prior to randomization, written informed consent was obtained
from all participants, assuring them of their voluntary participation and
the option to withdraw from the study at any time.
2.3. Procedure
The study recruited participants through referrals from rheumatol-
ogists at the HUVH Specialized Unit for Central Sensitivity Syndromes
and by contacting patients who had visited the Unit within the past six
months. Telephone assessments by project physiotherapist identied
eligible participants, who were then asked for written consent to
participate in the study. Remote assessments of clinical and health ser-
vices use were conducted, with a research assistant ensuring that par-
ticipants understood the self-reported measures. Participants were
randomly assigned to study groups following CONSORT guidelines
(Schulz et al. 2010) using a computer program (1:1:1 ratio). Adminis-
trative staff communicated group assignments to participants via email.
For those eligible for the biomarker substudy, a follow-up appointment
was scheduled 3–5 days after the initial evaluation to collect blood
samples, aiming to reach the required number of 40 subjects per arm.
Fasting blood draws occurred within a specied time slot (8–10 a.m.) to
minimize circadian variability. Participants were advised to avoid anti-
inammatory medications for 72 h beforehand to reduce the impact of
medication on the study results. The same blood collection procedure
was used after the intervention following the prespecied time interval.
The study withheld group assignment from evaluators and asked
participants not to disclose their treatment. Due to its non-
pharmacological nature, treatment assignment could not be blinded
for participants nor therapists.
2.4. Study arms
2.4.1. TAU
While there is not a universally accepted treatment for FM, within
the framework of Spanish healthcare, the TAU for FM primarily involves
pharmacological interventions tailored to the individual symptom pro-
le of each patient. Additionally, it typically includes recommendations
for aerobic physical exercise adapted to each individual’s limitations.
Participants allocated to this treatment group were given the option to
join either the FIBRO-On or FIBRO-Out programs upon completion of
the trial.
2.4.2. Fibro-On
Developed from the FIBROWALK program (Serrat et al. 2020), it is
designed to complement traditional treatments for FM by offering a
comprehensive online intervention. It includes explanations and
guidelines for practicing PSE, therapeutic physical exercise, CBT, and
mindfulness, delivered through videos and slide presentations by the
rst author (MS), an experienced physical therapist and health psy-
chologist. Each online session lasted 60 min, but participants were
recommended to spend 120 min completing the exercises. Weekly,
participants received links to the module videos via Adherence was
monitored through weekly questionnaires, with direct support from the
supervising therapist (MS) via mail or telephone for those who did not
respond or reported problems. Initiated during connement by COVID-
19, FIBRO-On has been integrated into routine clinical practice at the
Central Sensitivity Syndromes Specialized Unit of HUVH (Serrat et al.
2021a, 2022b).
2.4.3. Fibro-Out
The FIBRO-Out program is an outdoor version of the FIBROWALK
program (Serrat et al. 2020) as a complementary element to TAU. It
consisted of 12 weekly group sessions held in the “Parc del Cargol”, a
green area near HUVH in Barcelona, with 18–20 participants each.
Identical to FIBRO-On, it integrates PSE, therapeutic physical exercise,
CBT and mindfulness. The PSE, which is based on the “Explaining Pain”
program by Butler and Moseley (2010), whose content uses examples,
images and metaphors to deepen comprehension (Nijs et al. 2011).
Psychological aspects are based on CBT with the aim of reshaping the
understanding of pain, reducing pain catastrophizing, improving
emotional regulation and sleep quality, and developing coping strategies
(Moix and Kovacs, 2009). Mindfulness exercises, based on Mindfulness-
Based Stress Reduction (MBSR) program (Kabat-Zinn, 2013), aim to
train attention to the present experience, fostering a non-evaluative
attitude.
The sessions were divided into two blocks: PSE and therapeutic ex-
ercise led by a physiotherapist (1 h duration), followed by CBT and
mindfulness training conducted by a health psychologist (1 h duration).
Each outdoor session began with a discussion of key issues and a review
of previous concepts. Four pairs of therapists, trained in CBT and
physiotherapy for FM or chronic pain, managed the groups after
receiving specialized training and participating as co-therapists in a pilot
program. Weekly meetings with the clinical study leader ensured con-
sistency and addressed challenges. Treatment delity in the FIBRO-Out
arm was monitored through assessments by independent experts, and
participants were advised to continue their usual medication throughout
the study.
2.5. Outcome variables
2.5.1. Study measures
−Sociodemographic questionnaire: gender, date of birth, marital sta-
tus, cohabitation, educational level, and employment status.
−Clinical data: years with FM, comorbidity with other diagnosed
medical/psychiatric conditions, current medication, body mass
index (BMI).
2.5.2. Primary outcome measure
The Revised Fibromyalgia Impact Questionnaire (FIQR; Bennet et al.
2009) is a 21-item questionnaire (0–10 scale) that assesses the di-
mensions of physical dysfunction, the overall impact of FM and severity
of the symptoms (i.e., pain, energy, stiffness, sleep quality, depression,
S. Ferr´
es et al. Brain Behavior and Immunity 125 (2025) 184–197
186
memory issues, anxiety, allodynia, balance problems, and increased
sensitivity to noises, lights, smells, or temperatures), and is used to
measure the impact of FM over the past week. This questionnaire is
currently considered the “gold standard”for assessing functional
impairment in individuals with FM. A total score for FIQR ranging from
0 to 100 can be obtained by adding the 3 subscales, with higher scores
indicating greater FM severity. The Spanish version of the FIQR has an
excellent internal consistency (
α
=0.91–0.95) (Luciano et al. 2013).
2.5.3. Secondary outcome measures
Visual-analogue scale of perceived pain (VAS-Pain; Serrano-Atero et al.
2002) in which patients indicate their pain during the last week on a 10
cm line (0 =No pain, 10 =Unbearable pain).
Visual-analogue scale of perceived energy/fatigue (VAS-Fatigue;
Serrano-Atero et al. 2002) in which patients indicate their fatigue during
the last week on a 10 cm line (0 =Lots of energy, 10 =No energy).
The Hospital Anxiety and Depression Scale (HADS; Zigmond and
Snaith, 1983) is used to quantify the severity of anxiety and depression
symptoms. It consists of two dimensions (anxiety and depression) of 7
items each responding on a Likert scale of 4 points. Total scores of each
scale (HADS-A and HADS-D) range from 0 to 21, where higher scores
indicate greater symptom severity. The Spanish version of the HADS has
demonstrated satisfactory internal consistency for anxiety (
α
=0.83)
and depression (
α
=0.87) subscales in people with FM (Luciano et al.
2014).
The Physical Function Subscale of the Short Form-36 Health Survey (SF-
36; Alonso et al. 1995) is used to measure physical function. It comprises
a total of 10 items, which are answered on a Likert scale of 3 points.
Total scores are transformed and can range from 0 to 100, with higher
scores indicating better physical function. The Spanish version of the PF-
SF-36 shows adequate internal consistency (
α
=0.94).
The Tampa Scale for Kinesiophobia (TSK-11; Tkachuk and Harris,
2012) is a measure aimed at assessing fear of movement and comprises
11 items in a 4-point Likert scale with a total score ranging from 11 to
44, with higher scores indicating greater pain and fear of movement.
The Spanish version of the TSK-11 (G´
omez-P´
erez et al. 2011) has an
adequate internal consistency (
α
=0.79).
The FIBROWALK Fidelity Measure (FFM; Serrat et al. unpublished
manuscript) was employed to assess the extent to which therapists
adhered to the FIBROWALK protocol. This tool specically examines
therapists’delity to the FIBROWALK principles, protocols, and pre-
scribed methods, ensuring the therapy is delivered as intended. This
measure assessed therapists’adherence to FIBROWALK principles across
ve areas: general organization, therapist knowledge, personal skills,
therapy-related skills, and group management. Each of the 20 items was
rated on a 5-point Likert scale (0 =not met, 4 =fully met), with total
scores ranging from 0 to 20 with higher scores indicating greater
treatment delity. It is recommended that at least three out of the 12
sessions are evaluated: one within the rst four sessions, another in the
middle four sessions, and the last one during the nal four sessions.
2.6. Serum levels of immune-inammatory markers and BDNF
Blood samples were collected in vials that were centrifuged and the
resulting serum was stored frozen at −80 ◦C until analysis. All samples
(pre and post) were analyzed in a single analytical batch to reduce inter-
assay variability (approximately 15 %). Serum levels of the cytokines
and chemokines IL-6, CXCL8, IL-17A, IL-4 and IL-10, in addition to hs-
CRP, were assessed. For cytokine quantication, MerckMillipore Milli-
plex®reagents analyzed on a Luminex®platform were used. The highly
sensitive Human High Sensitivity T Cell multiplex kit (catalog number:
ME-HSTCMAG-28SK-05) was used. The hs-CRP was quantied by
turbidimetry on a Siemens Atellica autoanalyzer. BDNF levels were
assessed using an ELISA kit (reference SEA011Hu-96 T). Sample analysis
was performed by the Echevarne Laboratory. Detection concentration
ranges: IL-6 (0.73–10,000 pg/ml), CXCL8 (0.31–10,000 pg/ml), IL-17A
(2.93–20,000 pg/ml), IL-4 (7.32–10,000 pg/ml), IL-10 (1.46–40,000
pg/ml), hs-CRP (0.1–50 mg/L), and BDNF (31.2–2,000 pg/ml). Bio-
markers were assessed only at baseline and post-intervention for the
following reasons: a) evidence of changes in immune-inammatory
markers and BDNF levels after non-pharmacological interventions of
similar duration (Montero-Marin et al. 2019; P´
erez-Aranda et al. 2019;
Sanada et al. 2015); b) lower risk of sample loss (vs. assessment at 6
months); c) possibility of using baseline levels and pre-post change as a
mediator of clinical changes at 6-month follow-up; and d) budgetary
constraints.
2.7. Statistical analyses
Descriptive statistics were calculated for all variables and presented
as means (M) and standard deviations (SD) if continuous, or as absolute
numbers (n) and percentages (%) if categorical. Levene’s test was used
to evaluate the equality of variances of continuous variables and the
Kolmogorov-Smirnov test was used to test the normality and distribu-
tion of the samples. In those cases where biomarker concentrations were
below the detection threshold (31.4 % in IL-6, 1 % in CXCL8, 1 % in IL-
17a, 16.2 % in IL-4, 38.1 % in IL-10, 1.9 % in hs-CRP and 0 % in BDNF),
the detection limit value was assigned.
2.7.1. Analyses of short-term clinical effectiveness
The primary effectiveness analysis to assess the treatment effect on
FM was conducted using an intention-to-treat (ITT) approach, focusing
on changes in FIQR total scores (McCoy, 2017). The analysis used
restricted maximum likelihood (REML) mixed-effects linear regressions,
which are suitable for handling correlated repeated measures and pro-
vide more accurate variance estimates for small or unbalanced data sets
(Egbewale et al. 2014). This method did not require imputation for
missing data, as longitudinal mixed-model analysis can be performed
without it for any type of missing data (Twisk et al. 2013).
The analysis assessed the interaction between treatment groups and
time by calculating unstandardized regression coefcients (B) and 95 %
condence intervals (95 % CI) for these interactions at post-treatment.
Effect sizes for between-group differences were measured using
Cohen’s d. Effect sizes were calculated for the mean differences of
groups with unequal sample size within a pre-post-control design
(Morris, 2008), applying standard cut-off points for small (0.20), me-
dium (0.50) and large (0.80) effects. The same statistical approach was
applied to the secondary clinical endpoints.
2.7.2. Analyses of changes in biomarkers
The effect of the interventions on immune-inammatory markers
and BDNF levels was assessed with REML. Cytokine/chemokine, hs-
CRP, and BDNF values were natural log-transformed to normalize
skewed data distributions. Additionally, to assess the balance between
pro-inammatory and anti-inammatory responses, and to explore how
this balance might inuence treatment outcomes and guide strategies
for restoring immune homeostasis, inammatory balance indexes were
calculated following the approach of Andres-Rodriguez et al. (2019) and
Maes and Carvalho (2018). The calculated ratios included: IL6/IL4, IL6/
IL10, CXCL8/IL4, CXCL8/IL10, hsCRP/IL4, hsCRP/IL10, IL-17A/IL4,
and IL-17A/IL10. Since it has been reported that some medical treat-
ments, such as antidepressants, can affect some cytokine levels
(Hannestad et al. 2011), antidepressant status (0 =not taking, 1 =
taking) it was also included as a covariate in the REML analyses for all
clinical outcomes and biomarkers.
All analyses were conducted on an ITT basis and sensitivity analyses
were also conducted in a per-protocol approach or PP (i.e. including
only those participants who attended 9 or more sessions out of 12).
2.7.3. Predictive value of baseline biomarkers on the clinical response of
FIBROWALK
We compared sociodemographic and clinical characteristics and
S. Ferr´
es et al. Brain Behavior and Immunity 125 (2025) 184–197
187
biomarker values at baseline between responders and non-responders in
each of the FIBROWALK arms by computing t-tests or chi-square tests.
Furthermore, we examined the predictive value of baseline immune
biomarkers on the clinical effects of FIBROWALK, following the
approach by Judd et al. (2001). To carry this out, change scores
(calculated as post-treatment minus pre-treatment scores) for the
assessed clinical measures were analyzed in relation to baseline levels of
immune biomarkers and their indices using a stepwise approach. Soci-
odemographic and clinical variables (e.g., age, BMI, antidepressant use,
ISPS, and comorbidities such as CFS, depression, and anxiety) were
included in the rst step of the regression analyses and baseline
biomarker levels and related ratios in a second step, both steps using the
stepwise method. Negative beta values between baseline variables and
change scores should be interpreted in the sense that higher baseline
biomarker values predict greater clinical improvements for all clinical
measures except for the SF-36 scale, which should be interpreted
inversely.
All analyses were conducted with SPSS v26.0 and the signicance
level was established at
α
=0.05 (two-tailed).
3. Results
The owchart of the On&Out substudy (based on the consolidated
standards of reporting trials [CONSORT] recommendations) is displayed
in Fig. 1.
3.1. Demographic and baseline characteristics of the groups
A total of 120 individuals with FM (n=40 per group) were included
and randomized to the three treatment arms. Participants had a mean
age of 55 years. About 28 % were employed, 64 % were married or in a
stable relationship, 83 % were living with someone, and 78 % had at
least a high school education. 47 % of the participants reported some
level of disability, and 51 % had a comorbid diagnosis of chronic fatigue
syndrome. Participants showed high functional impairment, with a
mean FIQR score of 70 out of 100 (Table 1 and S1).
Fig. 1. Flowchart of the On&Out substudy.
S. Ferr´
es et al. Brain Behavior and Immunity 125 (2025) 184–197
188
3.2. Short-term effectiveness of the FIBRO-On and FIBRO-Out on clinical
measures
The therapists involved in the FIBRO-Out intervention achieved an
average FFM score of 18.42 out of 20 (SD =0.55), with individual scores
ranging from 17.95 to 19.46. The therapist tandems, consisting of a
psychologist and a physiotherapist, across the four FIBRO-Out groups
had an average FFM score of 18.75 out of 20 (SD =0.53), with tandem
averages varying between 18.07 and 19.17.
Signicant between-group differences were observed in clinical
variables at post-treatment (Table 2). Compared to TAU, signicantly
larger decreases in FIQR scores were found in the FIBRO-Out group
(medium effect size) and in the FIBRO-On group (small effect size). The
following secondary outcome variables showed signicantly larger im-
provements in the FIBRO-Out group vs TAU: VAS-Pain, VAS-Fatigue and
HADS-D and TSK (medium effect sizes). No signicant differences
between FIBRO-Out and TAU were found for the HADS-A and the SF-36.
When comparing FIBRO-On and TAU, signicant effects were only
found in the secondary outcome variable of TSK (large effect size). No
signicant differences were found regarding any clinical variable when
comparing both active treatment arms (all p >0.05).
When replicating the analyses on PP approach we found similar
treatment effects to those observed in ITT analyses (Supplementary
Table S2). Unlike the ITT approach, we found that patients in the FIBRO-
On group when compared to TAU presented greater reductions in VAS-
Fatigue scores (medium effect size) and that the FIBRO-Out arm showed
signicant differences in VAS-Pain scores (medium effect size)
compared to TAU.
3.3. Effects of the FIBRO-On and FIBRO-Out on serum biomarkers
Compared to TAU, the FIBRO-Out group showed greater reductions
in IL-6 levels (small effect size), while the FIBRO-On group exhibited
only an attenuation of the IL-6 increase over time (relative to TAU).
Additionally, larger reductions in IL-10 values were observed in the
FIBRO-Out group compared to TAU (small effect size). No other signif-
icant differences were found for any other immune-inammatory
biomarker nor BDNF levels (all p>0.05). For more details, see Table 3.
Regarding supplementary analyses on inammatory indexes, sig-
nicant effects of FIBRO-Out were found in the IL-6/IL-4 ratio where
FIBRO-Out produced a decrease in IL-6/IL-4, which is signicantly
different from the pattern of change seen in the TAU group, where we
observed an increase in this ratio. Larger decreases in CXCL8/IL-4 ratio
were also found in the FIBRO-On arm compared to FIBRO-Out
(Table S3). We added as a covariable the antidepressant status and
found no effect of antidepressants on our ndings (Table S4 and S5).
When replicating the analyses in the PP approach, differences were
found in the results from those observed in the ITT analyses (see
Table S6). Participants in the FIBRO-Out group showed signicantly
greater decreases in IL-4 levels compared to FIBRO-On (small effect
size); whereas IL-10 levels in the FIBRO-Out group (vs. TAU), showed
only a trend towards statistical signicance (p =0.051). It was also
observed that changes in the IL-6/IL-4 ratio in the FIBRO-Out group
were no longer signicant relative to TAU. Participants in the FIBRO-On
group (vs TAU) showed signicantly larger decreases in the CXCL8/IL-4
ratio (small effect size). This effect was not observed in the FIBRO-Out
group. Decreases in the IL-17A/IL-4 ratio were signicantly larger in
the FIBRO-On group compared to FIBRO-Out (small effect size)
(Table S7). No other signicant differences were found for any other
biomarker. Table 4 summarizes the main effects of the ITT and PP an-
alyses, highlighting signicant results and trends in clinical and
biomarker outcomes.
3.4. Predictive role of immune-inammatory biomarkers on response to
treatments
While FIBRO-On and FIBRO-Out demonstrated an overall improve-
ment in the clinical symptomatology of study participants, a consider-
able proportion of participants did not show a clinically relevant
response to treatment. Specically, 62.5 % of the FIBRO-On group and
60.6 % of the FIBRO-Out group failed to show a 20 % or higher reduc-
tion in the FIQR score. Subsequently, we categorized participants into
responders and non-responders and conducted analyses looking for
potential baseline differences between these two groups (refer to
Tables S8 and S9). Notably, no statistically signicant differences be-
tween responders and non-responders emerged in sociodemographic nor
baseline clinical variables, session attendance, nor baseline levels in
evaluated biomarkers (all p>0.05).
However, regression analyses indicated that in FIBRO-Out, higher
levels of BDNF before treatment predicted greater improvement in the
HADS-A and TSK; higher baseline levels of hs-CRP also predicted larger
improvements in the TSK (Table S10). In FIBRO-On, higher levels of IL-4
Table 1
Demographic and baseline clinical characteristics by treatment groups.
TAU FIBRO-Out FIBRO-On
(n¼35) (n¼37) (n¼33)
Age (years), M(SD)57.89 (9.71) 55.49 (10.16) 54.73 (10.33)
BMI, M(SD)29.40 (5.57) 28.24 (5.38) 27.71 (5.69)
ISPS, M(SD)17.86 (10.88) 10.43 (9.88) 13.15 (9.22)
With CFS, n(%) 18 (51.40) 17 (45.90) 19 (57.60)
Civil Status, n(%)
Single 3 (8.60) 3 (8.10) 5 (15.20)
Married/Living with partner 23 (65.70) 22 (59.50) 22 (66.70)
Divorced/Separated 7 (20.00) 6 (16.20) 4 (12.20)
Widow 2 (5.70) 6 (16.20) 2 (6.10)
Not living Alone, n(%) 28 (80.00) 31 (83.80) 28 (84.80)
Educational Level, n(%)
Without Studies 2 (5.70) 0 (0.00) 0 (0.00)
Primary Studies not completed 4 (11.40) 3 (8.10) 2 (6.10)
Primary Studies 9 (25.70) 8 (21.60) 6 (18.20)
Secondary Studies 15 (42.90) 17 (45.90) 16 (48.50)
University 4 (11.40) 8 (21.60) 7 (21.20)
Other 1 (2.90) 1 (2.70)) 2 (6.10)
Employment Situation, n(%)
Homemaker 4 (11.40) 3 (8.10) 3 (9.10)
Active 9 (25.70) 8 (21.60) 12 (36.40)
On leave 5 (14.30) 11 (29.70) 7 (21.20)
Unemployed with allowance 2 (5.70) 2 (5.40) 0 (0.00)
Unemployed without allowance 1 (2.90) 3 (8.10) 2 (6.10)
Retired/Pensioner 8 (22.90) 3 (8.10) 3 (9.10)
Temporary work disability 2 (5.70) 1 (2.70) 3 (9.10)
Other 4 (11.40) 6 (16.20) 3 (9.10)
Medication, n(%)
Analgesics 31(88.60) 32 (86.50) 27 (81.80)
Anticonvulsants 11 (31.40) 7 (18.90) 6 (18.20)
Antidepressants 30 (85.70) 30 (81.10) 16 (48.50)
Other 29 (82.90) 29 (78.40) 25 (75.80)
Incapacity certicate, n(%)
No 16 (45.70) 23 (62.20) 17 (51.50)
Less than 33 % 1 (2.90) 0 (0.00) 1 (3.00)
Between 33 % and 66 % 12 (34.30) 12 (32.40) 9 (27.30)
More than 66 % 6 (17.10) 2 (5.40) 4 (12.10)
Clinical variables, M(SD)
FIQR (0–100) 69.24 (18.48) 71.17 (14.80) 70.63 (15.99)
VAS-Pain (0–10) 7.89 (1.92) 7.95 (1.45) 7.73 (1.61)
VAS-Fatigue (0–10) 7.26 (2.74) 7.73 (2.02) 7.91 (1.96)
HADS-Anxiety (0–21) 14.49 (4.12) 12.35 (3.87) 12.36 (5.04)
HADS-Depression (0–21) 11.97 (4.20) 12.32 (3.71) 11.58 (4.13)
SF-36 (0–100) 30.86 (23.31) 33.65 (21.49) 33.76 (21.53)
TSK-11 (11–44) 31.26 (8.63) 28.89 (7.91) 32.64 (7.70)
Note: No comparison was found to be statistically signicant (p≦0.05). BMI,
Body Mass Index; CFS, Chronic Fatigue Syndrome; FIBRO-On, FIBROWALK
Online; FIBRO-Out, FIBROWALK Outdoor; FIQR, Revised Fibromyalgia Impact
Questionnaire; HADS-A and HADS-D, Hospital Anxiety and Depression Scale;
ISPS, Illness Self-Perceived Start; M, Mean; SF-36, Physical function subscale of
the Short Form-36 Health Survey; TAU, treatment-as-usual; TSK-11, Tampa
Scale for Kinesiophobia; VAS-Fatigue, Visual-analogue scale of perceived en-
ergy/fatigue; VAS-Pain, Visual-analogue scale of perceived pain.
S. Ferr´
es et al. Brain Behavior and Immunity 125 (2025) 184–197
189
Table 2
Descriptive Statistics and Between-group Analysis for Primary and Secondary Measures (ITT Approach).
TAU
(n¼
35)
Mean
(SD)
TAU þ
FIBRO-
On
(n¼33)
Mean
(SD)
TAU þ
FIBRO-
Out
(n¼37)
Mean
(SD)
TAU vs TAU þFIBRO-On TAU vs TAU þFIBRO-Out TAU þFIBRO-On vs
TAU þFIBRO-Out
F p d t (p) β(95 %
CI)
d t (p) β(95 %
CI)
d t (p) β(95 %
CI)
Primary
outcome
FIQR
(0–100) *
3.848 0.024
Baseline n=35
69.24
(18.48)
n=33
70.63
(15.99)
n=37
71.17
(14.80)
Post-
Treatment
n=34
67.57
(18.58)
n=32
61.36
(18.83)
n=33
61.09
(14.62)
0.43 ¡2.04
(0.044)
−7.25
(−14.32
to
−0.19)
0.50 ¡2.64
(0.010)
−9.23
(−16.16
to
−2.30)
0.05 −0.55
(0.580)
−1.97
(−9.01
to
5.06)
Secondary
outcomes
VAS-Pain
(0–10) *
2.924 0.058
Baseline n=35
7.89
(1.92)
n=33
7.73
(1.61)
n=37
7.95
(1.45)
Post-
Treatment
n=34
7.59
(2.46)
n=32
6.97
(1.86)
n=33
6.58
(1.70)
0.26 −0.80
(0.426)
−0.37
(−1.28
to
0.54)
0.63 ¡2.38
(0.019)
−1.07
(−1.97
to
−0.17)
0.36 −1.54
(0.126)
−0.70
(−1.61
to
0.20)
VAS-
Fatigue
(0–10) *
2.671 0.074
Baseline n=35
7.26
(2.73)
n=33
7.91
(1.95)
n=37
7.73
(2.02)
Post-
Treatment
n=34
7.79
(1.88)
n=32
7.19
(2.00)
n=33
6.97
(1.82)
0.52 −1.89
(0.061)
−1.22
(−2.49
to
0.05)
0.53 ¡2.08
(0.039)
−1.31
(−2.56
to
−0.06)
0.02 −0.15
(0.880)
−0.10
(−1.36
to
1.16)
HADS-A
(0–21) *
0.272 0.763
Baseline n=35
14.49
(4.12)
n=33
12.36
(5.04)
n=37
12.35
(3.86)
Post-
Treatment
n=34
14.03
(4.27)
n=32
11.16
(5.09)
n=33
11.61
(3.99)
0.16 −0.69
(0.487)
−0.57
(−2.18
to
1.04)
0.07 −0.13
(0.891)
−0.11
(−1.69
to
1.47)
−0.10 0.56
(0.574)
0.46
(−1.15
to
2.07)
HADS-D
(0–21) *
4.213 0.017
Baseline n=35
11.97
(4.20)
n=33
11.58
(4.13)
n=37
12.32
(3.71)
Post-
Treatment
n=34
12.68
(4.82)
n=32
10.71
(4.63)
n=33
10.45
(4.29)
0.38 −1.63
(0.105)
−1.36
(−3.02
to
0.29)
0.65 ¡2,89
(0.005)
−2.37
(−4.00
to
−0.74)
0.25 −1.20
(0.230)
−1.00
(−2.65
to
0.64)
SF-36
(0–100) *
0.522 0.595
Baseline n=35
30.86
(23.31)
n=33
33.79
(20.08)
n=37
33.65
(21.49)
Post-
Treatment
n=34
32.35
(23.07)
n=32
38.91
(23.51)
n=33
34.24
(20.70)
−0.17 0.74
(0.459)
2.95
(−4.93
to
10.84)
0.04 −0.24
(0.809)
−1.89
(−8.68
to
6.79)
0.22 −0.98
(0.327)
−3.90
(−11.76
to
3.95)
TSK-11
(11–44) *
11.123 (<0.001)
Baseline n=35
31.26
(8.63)
n=33
32.64
(7.70)
n=37
28.89
(7.91)
(continued on next page)
S. Ferr´
es et al. Brain Behavior and Immunity 125 (2025) 184–197
190
predicted lesser improvements in FIQR scores; greater levels of BDNF
predicted larger improvements in the SF-36; nally, higher values in the
CXCL8/IL-10 ratio predicted greater improvements in VAS-Pain and
VAS-Fatigue (Supplementary Table S11).
4. Discussion
In line with previous ndings on the short-term effectiveness of
FIBROWALK in inpatient (Serrat et al. 2021b), outdoor (Serrat et al.
2020), and online (Serrat et al. 2021a, 2022b) settings, our study
demonstrated that both FIBRO-On and FIBRO-Out effectively reduced
functional impairment and kinesiophobia in FM. Additionally, we found
both approaches to be equally effective in improving these core out-
comes in FM, with the outdoor version of FIBROWALK exhibiting
broader clinical effects compared to the online format, as it also allevi-
ated pain, fatigue, and depressive symptoms relative to TAU, whereas
the online format did not. In this regard, nature-based therapeutic ap-
proaches such as outdoor FIBROWALK (Serrat et al. 2020) have been
previously shown to be useful in improving mental health in different
clinical conditions (Trøstrup et al. 2019), including chronic pain con-
ditions and FM in particular (L´
opez-Pousa et al., 2015;Serrat et al. 2020;
Stanhope et al. 2020). Similarly, studies such as White et al. (2022) and
Serrat et al. (2021a) have highlighted the efcacy of online therapy in
chronic pain populations, including FM.
Surprisingly, despite previous ndings (Serrat et al. 2020, 2021a),
the FIBROWALK interventions in the present study did not signicantly
impact anxiety symptoms or physical function compared to TAU, with
very small effect sizes observed. Several factors may explain these re-
sults, including a smaller sample size, greater variability in physical
function scores, and lower comorbidity with CFS compared to prior
trials. Additionally, the use of linear mixed models, which tend to yield
more conservative estimates, may have also contributed to the lack of
signicant ndings.
Regarding the effects of FIBROWALK on serum biomarkers, we
observed signicant reductions in both IL-6 and IL-10 levels in the
FIBRO-Out group compared to TAU, indicating a potential immune-
inammatory modulation. While previous studies have reported re-
ductions in serum levels of CXCL8 and IL-6 in individuals with FM
following physical activity, multidisciplinary interventions, or dietary
changes (Sanada et al., 2015), our study found a signicant effect on IL-6
but not on CXCL8. IL-6 is often associated with a pro-inammatory
response, as it stimulates the immune system and the production of
acute-phase proteins; however, it also plays an anti-inammatory role
by inducing cytokine antagonists and supporting neuronal regeneration
(Maes et al., 2016; Raison et al., 2018). The dual nature of IL-6 becomes
evident during exercise, where it is released as a myokine from con-
tracting muscles, modulating the immune response with anti-
inammatory effects (Docherty et al., 2022; Nash et al., 2023; Hong-
Baik et al., 2023). Importantly, the impact of IL-6 varies depending on
the nature of the exercise: acute bouts tend to trigger short-term in-
creases, while regular, sustained physical activity leads to long-term
reductions (Docherty et al., 2022; Nash et al., 2023). In our study, the
FIBRO-Out group demonstrated a decrease in IL-6, while the FIBRO-On
group showed a slight increase. This difference may be explained by the
higher level of supervision and support provided in the FIBRO-Out
intervention, which likely encouraged greater adherence to regular
home-based exercises. Consequently, this more consistent physical ac-
tivity may have contributed to the observed reduction in IL-6 levels,
supporting the idea that regular exercise is more effective than acute
exercise in lowering IL-6 over time.
On the other hand, IL-10 plays a primarily anti-inammatory role,
suppressing pro-inammatory cytokines, modulating the activity of
various immune cells, and inhibiting antigen presentation, being key to
immune balance through the Compensatory Immune Regulatory Reex
System (Maes and Carvalho, 2018). The uctuation of anti-
inammatory cytokines, such as IL-10, in conjunction with pro-
inammatory ones (such as IL-6 in this instance), to regulate immu-
nity (Prather et al. 2007), may also contribute to the observed reduction
in IL-10 levels.
Interestingly, participants in the FIBRO-On group showed a less
pronounced increase in IL-6 levels compared to those in the TAU group;
these ndings may suggest a buffering effect of the intervention against
the natural progression toward a more pro-inammatory status, often
associated with aging (see Andr´
es-Rodríguez et al., 2020 where IL-6 was
found to be positively correlated with age). Considering that the
FIBROWALK intervention includes CBT and mindfulness training
alongside therapeutic physical exercise, the effects of CBT on circulating
pro-inammatory cytokines have been studied in FM patients with
studies showing signicant decreases in serum IL-6 and CXCL8 levels
compared to a wait-list control group (e.g., Zabihiyeganeh et al. 2019).
There is also existing evidence suggesting that mindfulness-based in-
terventions could also mitigate the inclination toward a more pro-
inammatory status in FM over time, similarly as we found in the
FIBRO-On arm (Andr´
es-Rodríguez et al. 2019). However, in Andr´
es-
Rodriguez et al.’s study (2019), the preventive effect primarily targeted
reducing the tendency toward a decrease in IL-10 observed in the TAU
group. In contrast, in our study, the focus was on IL-6, aiming to prevent
its increase in the FIBROWALK arm compared to TAU.
Consistently with the main results regarding individual cytokines/
chemokines, we found signicant differences in pro-inammatory/anti-
inammatory ratios between groups, with the outdoor intervention
showing larger decreases in IL-6/IL-4 ratio compared to TAU suggesting
a more balanced pro/anti-inammatory status after the intervention.
Interestingly, although no statistical differences were found between
active arms regarding clinical variables, greater reductions in CXCL8/IL-
4 ratio were found in the FIBRO-On intervention compared to FIBRO-
Table 2 (continued )
TAU
(n¼
35)
Mean
(SD)
TAU þ
FIBRO-
On
(n¼33)
Mean
(SD)
TAU þ
FIBRO-
Out
(n¼37)
Mean
(SD)
TAU vs TAU þFIBRO-On TAU vs TAU þFIBRO-Out TAU þFIBRO-On vs
TAU þFIBRO-Out
Post-
Treatment
n=34
29.82
(9.43)
n=32
24.22
(7.71)
n=33
23.18
(6.56)
0.84 ¡4.60
(<
0.001)
−6.66
(−9.54
to
−3.79)
0.51 ¡3.17
(0.002)
−4.51
(−7.33
to
−1.69)
−0.34 1.49
(0.139)
2.15
(−0.71
to
5.01)
Note. The baseline level of the variable was controlled. Mand SD are not adjusted. When antidepressants are taken as a covariate there is no signicant difference. *The
baseline level of the variable and study waves are signicant covariates in the model. CI, condence interval; d, Cohen’sdas an effect size measure; FIBRO-On,
FIBROWALK Online; FIBRO-Out, FIBROWALK Outdoor; FIQR, Revised Fibromyalgia Impact Questionnaire; HADS-A and HADS-D, Hospital Anxiety and Depression
Scale; ITT, intention-to-treat; β, regression coefcients; SF-36, Physical function subscale of the Short Form-36 Health Survey; TAU, treatment-as-usual; TSK-11, Tampa
Scale for Kinesiophobia; VAS-Fatigue, Visual-analogue scale of perceived energy/fatigue; VAS-Pain, Visual-analogue scale of perceived pain.
S. Ferr´
es et al. Brain Behavior and Immunity 125 (2025) 184–197
191
Table 3
Descriptive Statistics and Between-group Analysis for Biomarkers (ITT Approach).
TAU
(n¼
35)
Mean
(SD)
TAU þ
FIBRO-
On
(n¼33)
Mean
(SD)
TAU þ
FIBRO-
Out
(n¼37)
Mean
(SD)
TAU vs TAU þFIBRO-On TAU vs TAU þFIBRO-Out TAU þFIBRO-On vs
TAU þFIBRO-Out
F p d t (p) β(95 %
CI)
d t (p) β(95 %
CI)
d t (p) β(95 %
CI)
Inammatory
markers
IL-6 * (pg/ml) 4.673 0.011
Baseline n=35
0.90
(1.21)
n=33
1.06
(1.33)
n=37
1.20
(1.40)
Post-Treatment n=29
1.04
(1.10)
n=31
1.10
(1.34)
n=31
1.02
(1.24)
0.08 ¡2,03
(0.045)
−0.27
(−0.55
to
−0.01)
0.24 ¡3.00
(0.003)
−0.40
(−0.67
to
−0.13)
0.16 −0.94
(0.346)
−0.12
(−0.39
to
0.14)
CXCL8 * (pg/
ml)
0.325 0.723
Baseline n=35
2.45
(0.97)
n=33
2.67
(0.72)
n=37
2.76
(0.82)
Post-Treatment n=29
2.46
(0.84)
n=31
2.68
(0.74)
n=31
2.81
(0.65)
0.00 −0.59
(0.557)
−0.06
(−0.27
to
0.14)
−0.04 0.17
(0.861)
0.02
(−0.18
to
0.22)
−0.05 0.77
(0.441)
0.08
(−0.12
to
0.28)
IL-17A * (pg/
ml)
0.577 0.564
Baseline n=35
4,03
(0,70)
n=33
3,76
(0,82)
n=37
4,04
(0,48)
Post-Treatment n=29
4,00
(0,61)
n=31
3,79
(0,85)
n=31
4,08
(0,44)
−0.08 −0.21
(0.831)
−0.01
(−0.14
to
0.11)
−0.12 −1.01
(0.314)
−0.07
(−0.20
to
0.06)
−0.02 −0.80
(0.421)
−0.05
(−0.18
to
0.07)
IL-4 * (pg/ml) 2.049 0.134
Baseline n=35
3,72
(1,23)
n=33
3,60
(1,53)
n=37
4,07
(1,41)
Post-Treatment n=29
3,83
(1,30)
n=31
3,76
(1,52)
n=31
3,99
(1,25)
−0.04 0.58
(0.560)
0.07
(−0.17
to
0.31)
0.14 −1.37
(0.173)
−0.17
(−0.41
to
0.07)
0.16 −1.96
(0.052)
−0.24
(−0.48
to 0.00)
IL-10 * (pg/ml) 2.595 0.079
Baseline n=35
1,56
(1,30)
n=33
1,97
(1,63)
n=37
2,03
(1,59)
Post-Treatment n=29
1,75
(1,34)
n=31
2,03
(1,62)
n=31
1,78
(1,43)
0.09 −1.19
(0.235)
−0.23
(−0.61
to
0.15)
0.30 ¡2.27
(0.025)
−0.43
(−0.81
to
−0.05)
0.19 −1,07
(0.287)
−0.20
(−0.58
to
0.17)
hs-CRP * (mg/
L)
1.310 0.274
Baseline n=35
0,69
(0,82)
n=33
0,33
(1,13)
n=37
0,79
(1,13)
Post-Treatment n=29
0,73
(1,00)
n=31
0,12
(1,09)
n=31
0,84
(1,36)
0.25 −1.22
(0.222)
−0.23
(−0.60
to
0.14)
−0.01 0.27
(0.783)
0.05
(−0.32
to
0.42)
−0.23 1.52
(0.130)
0.28
(−0.08
to
0.65)
BDNF * (pg/ml) 0.789 0.453
Baseline n=35
4,40
(0,60)
n=33
4,23
(0,50)
n=37
4,37
(0,66)
Post-Treatment n=29
4,24
(0,45)
n=31
4,29
(0,55)
n=31
4,28
(0,27)
−0.39 1.18
(0.241)
0.22
(−0.15
to
0.60)
−0.11 0.20
(0.842)
0.03
(−0.33
to
0.40)
0.25 −0.99
(0.322)
−0.18
(−0.55
to
0.18)
Note. The baseline level of the variable was controlled. Mand SD are not adjusted. When antidepressants are taken as a covariate there is no signicant difference. *The
baseline level of the variable and study waves are signicant covariates in the model. BDNF, Brain-Derived Neurotrophic Factor; CI, condence interval; d, Cohen’sdas
an effect size measure; FIBRO-On, FIBROWALK Online; FIBRO-Out, FIBROWALK Outdoor; hs-CRP, high-sensitivity C-reactive protein; ITT, intention-to-treat; TAU,
treatment-as-usual; β, regression coefcients.
S. Ferr´
es et al. Brain Behavior and Immunity 125 (2025) 184–197
192
Out.
The ITT and PP analyses revealed some divergences in the impact of
FIBROWALK interventions on immune-inammatory biomarkers and
related ratios. A deactivation of the compensatory anti-inammatory
response (Maes and Carvalho, 2018) was observed in the FIBRO-Out
group, particularly when compared to FIBRO-On. In the PP analyses, a
signicant group x time effect was found for IL-4, with increases in
FIBRO-On and decreases in FIBRO-Out, suggesting decreased anti-
inammatory activity in the outdoor version of FIBROWALK. Addi-
tionally, the reduction in IL-10 levels in the FIBRO-Out group (vs TAU)
observed in the ITT analyses remained marginally signicant in the PP
analyses. The IL-6/IL-4 ratio reduction in FIBRO-Out (vs TAU) also
became non-signicant, while the IL-17A/IL-4 ratio increased in FIBRO-
Out (vs TAU) in the PP analyses. These ndings suggest that higher
adherence to the outdoor intervention, with increased exposure to nat-
ural environments and physical exercise, may be associated with a short-
term adaptive inammatory response, particularly through a reduction
in compensatory anti-inammatory activity (Maes and Carvalho, 2018;
Pernambuco et al., 2013; Nash et al., 2023). In this context, the reduc-
tion in IL-4 levels could also indicate a normalization of the overactive
Th2 response frequently observed in FM, where elevated IL-4 levels have
been reported (Maes and Carvalho, 2018; Andr´
es-Rodríguez et al.,
2020). Interestingly, IL-6 reductions in the FIBRO-Out group (vs TAU)
were also observed in the PP analyses, supporting the normalizing effect
of the intervention on the upregulated immune-inammatory pheno-
type characteristic of FM (Andr´
es-Rodríguez et al., 2020). The simulta-
neous reduction in IL-6 (vs TAU) and the decrease in IL-4 levels (vs
FIBRO-On) in the FIBRO-Out group may indicate a shift in immune
regulation, suggesting a rebalancing of Th1/Th2 responses and
normalization of immune activity in bromyalgia. This shift could be
driven by the long-term anti-inammatory effects of regular outdoor
physical activity (Andersen, Corazon and Stigsdotter, 2021; Nash et al.,
2023), potentially mitigating the immune dysregulation commonly
associated with chronic low-grade inammation (Andr´
es-Rodríguez
et al., 2020).
In contrast, FIBRO-On, despite having identical therapeutic content,
showed a decrease in the IL-17A/IL-4 ratio compared to FIBRO-Out and
a reduction in the CXCL8/IL-4 ratio (vs TAU and FIBRO-Out), suggesting
a decrease in inammatory activity and an increase in compensatory
anti-inammatory responses. These results point to a stronger engage-
ment of the compensatory anti-inammatory response system and a
counterbalancing of the immune-inammatory response system, which
Table 4
Summary of statistically signicant (or with a tendency) results (ITT and PP approaches).
Variable ITT p-value; d-value ITT Interpretation PP p-value;
d-value
PP Interpretation Comparison
FIQR (F-On vs TAU)
p=0.044; d¼0.43
(F-Out vs TAU)
p=0.010; d¼0.50
Reduction in F-On and F-Out (F-On vs TAU)
p¼0.014; d¼0.51
(F-Out vs TAU)
p¼0.007; d=0.63
Reduction in F-On and F-Out Similar results in PP and
ITT analyses.
VAS-Pain (F-Out vs TAU)
p¼0.019; d¼0.63
Reduction in F-Out (F-Out vs TAU)
p¼0.025; d=0.69
Reduction in F-Out Similar results in PP and
ITT analyses.
VAS-Fatigue (F-On vs TAU)
p=0.061; d =0.52
(F-Out vs TAU)
p¼0.039; d¼0.53
F-On reduction / F-Out Reduction (F-On vs TAU)
p¼0.030; d=0.61
(F-Out vs TAU)
p=0.058; d =0.52
Reduction in F-On / F-Out
reduction
PP partially conrms ITT
ndings.
HADS-D (F-Out vs TAU)
p¼0.005; d¼0.65
Reduction in F-Out (F-Out vs TAU)
p¼0.005; d¼0.69
Reduction in F-Out Similar results in PP and
ITT analyses.
TSK-11 (F-On vs TAU)
p¼<0.001; d¼0.84
(F-Out vs TAU) p¼
0.002; d=0.51
Reduction in F-On and F-Out (F-On vs TAU)
p¼<0.001; d=0.87
(F-Out vs TAU)
p¼0.001; d=0.58
Reduction in F-On and F-Out Similar results in PP and
ITT analyses.
IL-4 (F-On vs F-Out)
p=0.052; d =0.16
Increase in F-On / Reduction in F-
Out
(F-On vs F-Out) p¼
0.011; d=0.36
(F-Out vs TAU)
p=0.059; d =0.31
Increase in F-On / Reduction in F-
Out
PP partially conrms ITT
ndings.
IL-6 (F-On vs TAU)
p¼0.045; d=0.08
(F-Out vs TAU)
p¼0.003; d=0.24
Attenuated increase in F-On /
Reduction in F-Out
(F-On vs TAU)
p¼0.035; d=0.10
(F-Out vs TAU)
p¼0.016; d=0.30
Attenuated increase in F-On /
Reduction in F-Out
Similar results in PP and
ITT analyses.
IL-10 (F-Out vs TAU)
p¼0.025; d=0.30
Reduction in F-Out (F-Out vs TAU) p =0.051;
d=0.39
Reduction in F-Out PP partially conrms ITT
ndings.
Ratio IL-6/IL-
4
(F-Out vs TAU)
p¼0.028; d=-0.27
Reduction in F-Out n.s. N/A PP does not conrm ITT
ndings.
Ratio CXCL8/
IL-4
(F-On vs F-Out)
p¼0.021; d=0.36
Reduction in F-On / Increase in F-
Out
(F-On vs F-Out) p ¼
0.003; d=0.62
(F-On vs TAU) p ¼0.048;
d=-0.25
Reduction in F-On / Increase in F-
Out
PP partially conrms ITT
ndings.
Ratio CXCL8/
IL-10
(F-Out vs TAU) p =0.091;
d=0.26
Increase in F-Out n.s. N/A PP does not conrm ITT
ndings.
Ratio hs-CRP/
IL-4
n.s. N/A (F-On vs F-Out)
p=0.091; d =0.26
Reduction in F-On / Increase in F-
Out.
PP does not conrm ITT
ndings.
Ratio hs-CRP/
IL-10
(F-Out vs TAU)
p=0.078; d =0.20
Increase in F-Out (F-Out vs TAU)
p=0.073; d =0.26
Increase in F-Out Similar results in PP and
ITT analyses.
Ratio IL-17A/
IL-4
n.s. N/A (F-On vs F-Out)
p¼0.048; d=0.39
Reduction in F-On / Increase in F-
Out
PP does not conrm ITT
ndings.
Note: Only statistically signicant effects (p≤0.05) or showing a tendency towards signicance (0.05 <p≤0.10) are displayed. ITT, intention-to-treat; PP, per-
protocol; d, Cohen’sdas an effect size measure; F-On, FIBROWALK Online; F-Out, FIBROWALK Outdoor; TAU, treatment-as-usual; FIQR, Revised Fibromyalgia
Impact Questionnaire; VAS-Fatigue, Visual-analogue scale of perceived energy/fatigue; VAS-Pain, Visual-analogue scale of perceived pain; HADS-D, Hospital Anxiety
and Depression Scale; TSK-11, Tampa Scale for Kinesiophobia; hs-CRP, high-sensitivity C-reactive protein.
S. Ferr´
es et al. Brain Behavior and Immunity 125 (2025) 184–197
193
could indicate that FIBRO-On’s format fostered sustained anti-
inammatory effects, helping mitigate the inammatory symptoms
typically associated with FM. This may be linked to the therapeutic
benets of PSE (Shields et al., 2020), CBT (Montero-Marin et al., 2019),
mindfulness (Andr´
es-Rodríguez et al., 2019), and regular exercise (Nash
et al., 2023; Sanada et al., 2015).
These ndings suggest different immune-inammatory pathways
involved in outdoor and online FIBROWALK. They contribute to an
intriguing line of research exploring how contextual factors (e.g., nat-
ural environments) —beyond the therapeutic content itself —in which
therapies are conducted may inuence immune system functioning and
interact with therapeutic components, ultimately shaping the effects of
multicomponent interventions on both immune responses and clinical
outcomes.
In our study, no signicant changes in BDNF levels were observed in
FIBROWALK interventions compared to TAU. Nevertheless, in other
studies where non-multicomponent (without including physical exer-
cise) psychological interventions (e.g., Montero-Marin et al. 2019;
Sanabria-Mazo et al. 2020) were performed, reductions in BDNF were
observed. Since increased levels of BDNF have been found in FM
compared to healthy participants, and such increases have been asso-
ciated with a chronication of pain, due to the key role of BDNF in
various neuroplasticity processes (Nugraha et al. 2012), reductions re-
ported in previous studies in FM were found to be indicative of a
benecial effect of the interventions. However, we must bear in mind
that FM is a complex disease with high comorbidity with mental diseases
such as depression and the specic clinical characteristics of the study
sample may have a major effect on the evaluated biomarkers. In this
regard, some studies have suggested that depression may be associated
with lower BDNF levels (Cavaleri et al. 2023) and not higher levels as it
has been reported in FM as we commented above. In this regard, one
potential difference between our study and previous reports nding
signicant reductions in BDNF after psychotherapeutic interventions for
FM (e.g., Montero-Marin et al. 2019; Sanabria-Mazo et al. 2020) may lie
in the fact that our sample, recruited from a specialized treatment unit,
exhibited higher levels of depressive symptoms compared to the par-
ticipants in these randomized controlled trials, who were recruited from
primary care settings. Therefore, the effects of a multicomponent
intervention such as FIBROWALK, which includes both psychothera-
peutic (which may decrease BDNF levels) and exercise approaches, may
not be straightforward.
Another signicant nding of our study is that −in both outdoor and
online formats of the FIBROWALK- participants showing a more pro-
inammatory prole at baseline experimented larger clinical improve-
ments after intervention. More precisely, it was found that higher pre-
treatment hs-CRP values predicted improvement in kinesiophobia in the
FIBRO-Out group. In the FIBRO-On group, higher levels of IL-4 predicted
worsening functional impairment, while higher values of the CXCL8/IL-
10 index predicted improvement in pain and fatigue. Since IL-4 is a key
regulator in humoral and adaptive immunity, our results suggest that
individuals with a higher basal compensatory response would have less
improvement in functional impairment after FIBRO-On program. In this
regard, a meta-analysis found elevated levels of IL-4, as well as IL-6 and
IL17A, in individuals with FM compared to healthy controls, suggesting
altered homeostasis with elevated immune-inammatory and compen-
satory pathways (Andr´
es-Rodríguez et al. 2020). However, it is impor-
tant to note that other studies evaluating the effects of different
cognitive-behavioral therapies, also conducted in FM (e.g., Andr´
es-
Rodríguez et al. 2019; Lasselin et al. 2016), found contrary results. In
this regard, higher basal levels of CXCL8 attenuated the benecial effect
of a MBSR intervention on clinical symptomatology, including pain,
energy, stiffness, or sleep quality, suggesting that basal inammation
may hinder clinical response in that sample after this specic treatment
(Andr´
es-Rodríguez et al. 2019); similarly, higher levels of IL-6 and TNF-
α
before treatment were associated with lesser improvement in pain
intensity (in a 0–6 scale) and psychological inexibility (also assessed
with the PIPS) following behavioral treatment (Lasselin et al. 2016).
Since larger clinical improvements after interdisciplinary approaches
can be particularly found in individuals with worse baseline clinical
status (e.g., Worrel et al. 2001) and both IL-6 and CXCL8 have been
found to positively correlate with symptom severity in FM (Andr´
es-
Rodríguez et al. 2019), a potential explanation to our ndings could be
that patients with a more severe clinical prole at baseline (and also
displaying greater pro-inammatory status) would show greater im-
provements after multicomponent interventions. Additionally, differ-
ences between our study sample −with participants showing a more
impaired clinical prole- and those from the Andr´
es-Rodríguez et al.
(2019) and Lasselin et al. (2016) studies, may also be behind the dif-
ferences in the results of regression analyses.
Finally, our study revealed intriguing insights into the predictive
value of pre-treatment BDNF values. Specically, higher baseline BDNF
levels in the FIBRO-Out group were associated with larger improve-
ments in anxiety symptoms and kinesiophobia. This suggests that
elevated BDNF levels may serve as a potential biomarker for identifying
individuals who are more likely to benet from the FIBRO-Out inter-
vention in terms of anxiety management and reducing movement-
related fear. Conversely, in the FIBRO-On group, higher pre-treatment
BDNF values predicted greater improvements in physical function.
This nding implies that elevated BDNF levels might facilitate larger
physical function improvements following the FIBRO-On program.
These ndings may be explained by the fact that BDNF play a crucial
role in neuroplasticity and learning (Colucci-D’Amato et al. 2020) and,
potentially, fostering processes both unadaptive—such as those pro-
moting central sensitization (Nijs et al., 2015; Xiong et al. 2024)—and
those adaptive—such as those related to therapeutic inter-
ventions—may be more easily promoted in those individuals with
greater BDNF levels. These results underscore the potential role of BDNF
as a predictive biomarker, suggesting that its levels could help tailor
treatment approaches to individual patient needs, optimizing the ef-
cacy of FIBROWALK interventions for FM.
This study has an exploratory nature and therefore there is still a
need for further studies on how immune functions may play a role in the
response to other multicomponent interventions in individuals with FM.
Our ndings should be interpreted in light of the limitations of the
present study. Firstly, the sample size was rather small and, therefore,
our study was somewhat underpowered. Furthermore, fteen samples
were lost post-randomization due to errors in sample processing and
delayed identication of unmet inclusion criteria (e.g., autoimmune
diseases, COVID-19). It was not possible to increase the number of
participants to the initially proposed 40 per group due to budget limi-
tations and because the treatment had already started when these issues
were identied. Secondly, there was variability in adherence between
intervention groups, with higher adherence observed in the FIBRO-On
group (91 %) compared to the FIBRO-Out group (78 %). While sensi-
tivity analyses were conducted to address the impact of attrition and
non-adherence on the results, these differences may have inuenced the
outcomes. The FIBROWALK program includes physical therapeutic ex-
ercises that provide signicant benets for the physical and emotional
health of patients with FM, alleviating symptoms of anxiety and
depression and thus improving overall well-being (Serrat et al. 2020).
However, the effectiveness of the program may be limited by variability
in patients’levels of pain, fatigue, and functional capacity, making it
difcult to customize it to meet individual needs. The outdoor program
is group-based, accommodating up to 20 patients, while the online
program is self-administered, which complicates its personalization.
Nevertheless, participants can consult a therapist if they have questions
during the program. Furthermore, it is important to consider that, in
general, the risk of overexertion during aerobic and strength exercises
may exacerbate pain and fatigue, negatively affecting motivation
(Thieme et al. 2017; Perrot and Russell, 2014). An under-dosing or over-
dosing of exercise could have affected the rate and magnitude of the
observed clinical response, as well as changes in the studied biomarkers,
S. Ferr´
es et al. Brain Behavior and Immunity 125 (2025) 184–197
194
which could have signicant implications for the study results and their
clinical applicability (Serrat et al., 2021b). Future studies incorporating
objective measures of effort (e.g., with Fitbit or similar devices) could
help resolve this issue by providing more accurate and consistent
monitoring of exercise intensity. Thirdly, the lack of blinding of par-
ticipants and therapists is a common limitation in studies of non-
pharmacological interventions, which may introduce bias into the re-
ported outcomes. Additionally, there is a possibility of biased reporting
of adherence to the FIBRO-On program, which could affect the inter-
pretation of the results. Conducting the study in a specialized clinical
setting may have implications for the generalizability of the ndings.
While it may increase the practical relevance of the results, it could also
limit their applicability to other settings or populations. Furthermore,
the study was constrained by budget limitations, resulting in a reduced
number of biomarkers studied. Additionally, the evaluation of
biomarker levels was limited to serum samples, providing only indirect
insight into the effects of the interventions on the inammatory status in
FM.
5. Conclusions
This study forms part of a broader clinical trial assessing the long-
term effectiveness and cost-effectiveness of FIBROWALK interventions
(Serrat et al. 2022b). Our ndings indicate that both online and outdoor
FIBROWALK interventions not only enhance the clinical status of FM
patients but also exert an impact on immune-inammatory pathways
implicated in the syndrome. Notably, baseline levels of the examined
biomarkers and their indices were predictive of varying responses to the
treatments assessed. Thus, integrating immune-inammatory bio-
markers and BDNF serum levels into treatment protocols may help
distinguish patient proles with differing responses to interventions.
Our results underscore the importance of incorporating these bio-
markers into clinical practice, paving the way for tailored treatment
plans for FM individuals. Such personalized approaches hold promise for
improving efciency, cost-effectiveness, and utilization of healthcare
services in FM management (Carvalho et al. 2019; Lopresti, 2017; Thase,
2014).
Declaration of generative AI and AI-assisted technologies in the
writing process
During the preparation of this work the authors used ChatGPT to
improve some sentences of the manuscript in terms of grammar and
style. After using this tool/service, the authors reviewed and edited the
content as needed and take full responsibility for the content of the
publication.
CRediT authorship contribution statement
S`
onia Ferr´
es: Writing –review &editing, Writing –original draft,
Visualization, Methodology, Investigation, Formal analysis, Data cura-
tion. Mayte Serrat: Writing –review &editing, Writing –original draft,
Visualization, Validation, Supervision, Resources, Project administra-
tion, Methodology, Investigation, Funding acquisition, Conceptualiza-
tion. William Auer: Writing –review &editing, Visualization. Estíbaliz
Royuela-Colomer: Writing –original draft, Data curation. Míriam
Almirall: Writing –review &editing. Andrea Lizama-Lefno: Writing –
review &editing. Jo Nijs: Writing –review &editing. Michael Maes:
Writing –review &editing, Methodology. Juan V. Luciano: Writing –
review &editing, Writing –original draft, Visualization, Validation,
Supervision. Xavier Borr`
as: Writing –review &editing, Writing –
original draft, Visualization, Validation, Supervision, Methodology,
Investigation. Albert Feliu-Soler: Writing –review &editing, Writing –
original draft, Visualization, Validation, Supervision, Software, Re-
sources, Project administration, Methodology, Investigation, Funding
acquisition, Formal analysis, Data curation, Conceptualization.
Funding
The project has been funded in part by the Spanish Ministry for
Science and Innovation (MCIN) State R +D+I Program Oriented to the
Challenges of Society-MCIN/AEI/10.13039/501100011033- (ref. PID
2020-117667RA-I00) and co-nanced with European Union ERDF
funds. The authors are grateful to the CIBER of Epidemiology and Public
Health (CIBERESP; CB06/02/0046 &CB22/02/00052) for their sup-
port. The funding sources had no inuence on the design of the study,
data collection and analysis, or the writing of the manuscript.
Declaration of competing interest
The authors declare the following nancial interests/personal re-
lationships which may be considered as potential competing interests:
JN and the Vrije Universiteit Brussel received lecturing/teaching fees
from various professional associations and educational organizations.
JN authored Dutch books on pain science education and pain manage-
ment. The remaining authors declare that they have no conict of
interest.
Appendix A. Supplementary data
Supplementary data to this article can be found online at https://doi.
org/10.1016/j.bbi.2024.12.149.
Data availability
Data will be made available on request.
References
Alonso, J., Prieto, L., Ant´
o, J.M., 1995. The Spanish version of the SF-36 Health Survey
(the SF-36 Health Questionnaire): an instrument for measuring clinical outcomes
measuring clinical results. Med. Clin. (Barc) 104, 771–776.
Andersen, L., Corazon, S.S., Stigsdotter, U.K., 2021. Nature exposure and its effects on
immune system functioning: a systematic review. Int. J. Environ. Res. Public Health
18 (4), 1416.
Andr´
es-Rodríguez, L., Borr`
as, X., Feliu-Soler, A., P´
erez-Aranda, A., Rozadilla-Sacanell, A.,
Montero-Marin, J., Maes, M., Luciano, J.V., 2019. Immune-inammatory pathways
and clinical changes in bromyalgia patients treated with Mindfulness-Based Stress
Reduction (MBSR): a randomized, controlled clinical trial. Brain Behav.. Immun. 80,
109–119. https://doi.org/10.1016/j.bbi.2019.02.030.
Andr´
es-Rodríguez, L., Borr`
as, X., Feliu-Soler, A., P´
erez-Aranda, A., Angarita-Osorio, N.,
Moreno-Peral, P., Montero-Marin, J., García-Campayo, J., Carvalho, A.F., Maes, M.,
Luciano, J.V., 2020. Peripheral immune aberrations in bromyalgia: a systematic
review, meta-analysis and meta-regression. Brain Behav.. Immun. 87, 881–889.
https://doi.org/10.1016/j.bbi.2019.12.020.
B¨
ackryd, E., Tanum, L., Lind, A.L., Larsson, A., Gordh, T., 2017. Evidence of both
systemic inammation and neuroinammation in bromyalgia patients, as assessed
by a multiplex protein panel applied to the cerebrospinal uid and to plasma. J. Pain
Res. 515–525.
Butler, D.S., Moseley, G.L., 2010. Explaining pain. Noigroup Publications.
Carvalho, A.F., Fernandes, B.S., Brunoni, A.R., 2019. Mindfulness-based stress reduction
for bromyalgia: a step closer to precision psychiatry? Brain Behav.. Immun. 81,
8–9. https://doi.org/10.1016/j.bbi.2019.06.020.
Cavaleri, D., Moretti, F., Bartoccetti, A., Mauro, S., Crocamo, C., Carr`
a, G., Bartoli, F.,
2023. The role of BDNF in major depressive disorder, related clinical features, and
antidepressant treatment: Insight from meta-analyses. Neurosci. Biobehav. Rev. 149,
105159. https://doi.org/10.1016/j.neubiorev.2023.105159.
Colucci-D’Amato, L., Speranza, L., Volpicelli, F., 2020. Neurotrophic factor BDNF,
physiological functions and therapeutic potential in depression, neurodegeneration
and brain cancer. Int. J. Mol. Sci. 21, 7777. https://doi.org/10.3390/ijms21207777.
De Miquel, C.A., Campayo, J., Fl´
orez, M.T., Arguelles, J.M., Tarrio, E.B., Montoya, M.G.,
Martin, ´
A., Salio, A.M., Fuentes, J.V., Alberch, E.A., de la C´
amara, A.G., 2010.
Interdisciplinary consensus document for the treatment of bromyalgia. Actas Esp.
Psiquiatr. 38, 108–120.
de Tommaso, M., Vecchio, E., Nolano, M., 2022. The puzzle of bromyalgia between
central sensitization syndrome and small ber neuropathy: a narrative review on
neurophysiological and morphological evidence. Neurol. Sci. 43, 1667–1684.
https://doi.org/10.1007/s10072-021-05806-x.
Deitos, A., Duss´
an-Sarria, J. A., Souza, A.d, Medeiros, L., Tarragˆ
o, M.daG., Sehn, F.,
Chassot, M., Zanette, S., Schwertner, A., Fregni, F., Torres, I. L., Caumo, W., 2015.
Clinical value of serum neuroplasticity mediators in identifying the central
sensitivity syndrome in patients with chronic pain with and without structural
S. Ferr´
es et al. Brain Behavior and Immunity 125 (2025) 184–197
195
pathology. Clin J Pain. 31, 959–967. https://doi.org/10.1097/
AJP.0000000000000194.
Di-Bonaventura, S., Fern´
andez-Carnero, J., Matesanz-García, L., Arribas-Romano, A.,
Polli, A., Ferrer-Pe˜
na, R., 2023. Effect of different physical therapy interventions on
brain-derived neurotrophic factor levels in chronic musculoskeletal pain patients: a
systematic review. Life (basel). 13, 163. https://doi.org/10.3390/life13010163.
Docherty, S., Harley, R., McAuley, J.J., Crowe, L.A.N., Pedret, C., Kirwan, P.D.,
Siebert, S., Millar, N.L., 2022. The effect of exercise on cytokines: implications for
musculoskeletal health: a narrative review. BMC Sports Sci. Med. Rehabil. 14, 5.
https://doi.org/10.1186/s13102-022-00397-2.
Egbewale, B.E., Lewis, M., Sim, J., 2014. Bias, precision and statistical power of analysis
of covariance in the analysis of randomized trials with baseline imbalance: a
simulation study. BMC Med. Res. Method. 14, 49. https://doi.org/10.1186/1471-
2288-14-49.
Hannestad, J., DellaGioia, N., Bloch, M., 2011. The effect of antidepressant medication
treatment on serum levels of inammatory cytokines: a meta-analysis.
Neuropsychopharmacology 36, 2452–2459. https://doi.org/10.1038/
npp.2011.132.
H¨
auser, W., Bernardy, K., Arnold, B., Offenb¨
acher, M., Schiltenwolf, M., 2009. Efcacy of
multicomponent treatment in bromyalgia syndrome: a meta-analysis of
randomized controlled clinical trials. Arthritis Rheum. 61, 216–224. https://doi.org/
10.1002/art.24276.
H¨
auser, W., Ablin, J., Fitzcharles, M.A., Littlejohn, G., Luciano, J.V., Usui, C., Walitt, B.,
2015. Fibromyalgia. Nat. Rev. Dis. Primers. 1, 15022. https://doi.org/10.1038/
nrdp.2015.22.
Heidari, F., Afshari, M., Moosazadeh, M., 2017. Prevalence of bromyalgia in general
population and patients, a systematic review and meta-analysis. Rheumatol. Int. 37,
1527–1539. https://doi.org/10.1007/s00296-017-3725-2.
Hong-Baik, I., Úbeda-D’Ocasar, E., Cimadevilla-Fern´
andez-Pola, E., Jim´
enez-Díaz-
Benito, V., Herv´
as-P´
erez, J.P., 2023. The effects of non-pharmacological
interventions in bromyalgia: a systematic review and meta analysis of
predominants outcomes. Biomedicines. 11, 2367. https://doi.org/10.3390/
biomedicines11092367.
Judd, C.M., Kenny, D.A., McClelland, G.H., 2001. Estimating and testing mediation and
moderation in within-subject designs. Psychol. Methods 6, 115–134. https://doi.
org/10.1037/1082-989x.6.2.115.
Kabat-Zinn, J., 2013. Full catastrophe living, revised edition. Hachette UK.
Lasselin, J., Kemani, M.K., Kanstrup, M., Olsson, G.L., Axelsson, J., Andreasson, A.,
Lekander, M., Wicksell, R.K., 2016. Low-grade inammation may moderate the
effect of behavioral treatment for chronic pain in adults. J. Behav. Med. 39,
916–924. https://doi.org/10.1007/s10865-016-9769-z.
Li, L.W., Chew, A.M.K., Gunasekeran, D.V., 2020. Digital health for patients with chronic
pain during the COVID-19 pandemic. Br. J. Anaesth. 125, 657–660. https://doi.org/
10.1016/j.bja.2020.08.003.
Lichtenstein, A., Tiosano, S., Amital, H., 2018. The complexities of bromyalgia and its
comorbidities. Curr. Opin. Rheumatol. 30, 94–100. https://doi.org/10.1097/
BOR.0000000000000464.
L´
opez-Pousa, S., Bassets Pag`
es, G., Monserrat-Vila, S., de Gracia Blanco, M., Hidalgo
Colom´
e, J., Garre-Olmo, J., 2015. Sense of well-being in patients with bromyalgia:
aerobic exercise program in a mature forest—a pilot study. Evid. Based
Complementary Alternat. Med. 2015 (1), 614783.
Lopresti, A.L., 2017. Cognitive behaviour therapy and inammation: a systematic review
of its relationship and the potential implications for the treatment of depression.
Aust. N. Z. J. Psychiatry 51, 565–582. https://doi.org/10.1177/
0004867417701996.
Luciano, J.V., Aguado, J., Serrano-Blanco, A., Calandre, E.P., Rodriguez-Lopez, C.M.,
2013. Dimensionality, reliability, and validity of the revised bromyalgia impact
questionnaire in two Spanish samples. Arthritis Care Res. (Hoboken). 65,
1682–1689. https://doi.org/10.1002/acr.22034.
Luciano, J.V., Barrada, J.R., Aguado, J., Osma, J., García-Campayo, J., 2014. Bifactor
analysis and construct validity of the HADS: a cross-sectional and longitudinal study
in bromyalgia patients. Psychol. Assess. 26, 395–406. https://doi.org/10.1037/
a0035284.
Macfarlane, G.J., Kronisch, C., Dean, L.E., Atzeni, F., H¨
auser, W., Fluß, E., Choy, E.,
Kosek, E., Amris, K., Branco, J., Dincer, F., Leino-Arjas, P., Longley, K., McCarthy, G.
M., Makri, S., Perrot, S., Sarzi-Puttini, P., Taylor, A., Jones, G.T., 2017. EULAR
revised recommendations for the management of bromyalgia. Ann. Rheum. Dis. 76,
318–328. https://doi.org/10.1136/annrheumdis-2016-209724.
Maes, M., Carvalho, A.F., 2018. The compensatory immune-regulatory reex system
(CIRS) in depression and bipolar disorder. Mol. Neurobiol. 55, 8885–8903. https://
doi.org/10.1007/s12035-018-1016-x.
Maes, M., Nowak, G., Caso, J.R., Leza, J.C., Song, C., Kubera, M., Klein, H., Galecki, P.,
Noto, C., Glaab, E., Balling, R., Berk, M., 2016. Toward omics-based, systems
biomedicine, and path and drug discovery methodologies for depression-
inammation research. Mol. Neurobiol. 2927–2935. https://doi.org/10.1007/
s12035-015-9183-5.
McCoy, C.E., 2017. Understanding the Intention-to-treat principle in randomized
controlled trials. West J. Emerg. Med. 18, 1075–1078. https://doi.org/10.5811/
westjem.2017.8.35985.
Moix, J., Kovacs, F.M. (Eds.)., 2009. Manual del dolor. Tratamiento cognitivo-conductual
del dolor cr´
onico. Barcelona: Paid´
os.
Montero-Marin, J., Andr´
es-Rodríguez, L., Tops, M., Luciano, J.V., Navarro-Gil, M., Feliu-
Soler, A., L´
opez-Del-Hoyo, Y., Garcia-Campayo, J., 2019. Effects of attachment-based
compassion therapy (ABCT) on brain-derived neurotrophic factor and low-grade
inammation among bromyalgia patients: a randomized controlled trial. Sci. Rep.
9, 15639. https://doi.org/10.1038/s41598-019-52260-z.
Morris, S.B., 2008. Estimating effect sizes from pretest-posttest-control group designs.
Organ. Res. Methods 11, 364–386. https://doi.org/10.1177/1094428106291059.
Nash, D., Hughes, M.G., Butcher, L., Aicheler, R., Smith, P., Cullen, T., Webb, R., 2023.
IL-6 signaling in acute exercise and chronic training: Potential consequences for
health and athletic performance. Scand. J. Med. Sci. Sports 33, 4–19. https://doi.
org/10.1111/sms.14241.
Nijs, J., Paul van Wilgen, C., Van Oosterwijck, J., van Ittersum, M., Meeus, M., 2011.
How to explain central sensitization to patients with ’unexplained’chronic
musculoskeletal pain: practice guidelines. Man. Ther. 16, 413–418. https://doi.org/
10.1016/j.math.2011.04.005.
Nijs, J., Meeus, M., Versijpt, J., Moens, M., Bos, I., Knaepen, K., Meeusen, R., 2015. Brain-
derived neurotrophic factor as a driving force behind neuroplasticity in neuropathic
and central sensitization pain: a new therapeutic target? Expert Opin. Ther. Targets
19, 565–576. https://doi.org/10.1517/14728222.2014.994506.
Nüesch, E., H¨
auser, W., Bernardy, K., Barth, J., Jüni, P., 2013. Comparative efcacy of
pharmacological and non-pharmacological interventions in bromyalgia syndrome:
network meta-analysis. Ann. Rheum. Dis. 72, 955–962. https://doi.org/10.1136/
annrheumdis-2011-201249.
Nugraha, B., Karst, M., Engeli, S., Gutenbrunner, C., 2012. Brain-derived neurotrophic
factor and exercise in bromyalgia syndrome patients: a mini review. Rheumatol.
Int. 32, 2593–2599. https://doi.org/10.1007/s00296-011-2348-2.
P´
erez-Aranda, A., Feliu-Soler, A., Montero-Marín, J., García-Campayo, J., Andr´
es-
Rodríguez, L., Borr`
as, X., Rozadilla-Sacanell, A., Pe˜
narrubia-Maria, M.T., Angarita-
Osorio, N., McCracken, L.M., Luciano, J.V., 2019. A randomized controlled efcacy
trial of mindfulness-based stress reduction compared with an active control group
and usual care for bromyalgia: the EUDAIMON study. Pain 160, 2508–2523.
https://doi.org/10.1097/j.pain.0000000000001655.
Pernambuco, A.P., Schetino, L.P., Alvim, C.C., Murad, C.M., Viana, R.S., Carvalho, L.S.,
Reis, D.´
A., 2013. Increased levels of IL-17A in patients with bromyalgia. Clin. Exp.
Rheumatol. 31 (6 Suppl 79), S60–S63.
Perrot, S., Russell, I.J., 2014. More ubiquitous effects from non-pharmacologic than from
pharmacologic treatments for bromyalgia syndrome: a meta-analysis examining six
core symptoms. Eur. J. Pain 18, 1067–1080. https://doi.org/10.1002/ejp.564.
Polli, A., Ghosh, M., Bakusic, J., Ickmans, K., Monteyne, D., Velkeniers, B., Bekaert, B.,
Godderis, L., Nijs, J., 2020. DNA methylation and brain-derived neurotrophic factor
expression account for symptoms and widespread hyperalgesia in patients with
chronic fatigue syndrome and comorbid bromyalgia. Arthritis Rheumatol. 72,
1936–1944. https://doi.org/10.1002/art.41405.
Prather, A.A., Marsland, A.L., Muldoon, M.F., Manuck, S.B., 2007. Positive affective style
covaries with stimulated IL-6 and IL-10 production in a middle-aged community
sample. Brain Behav. Immun. 21, 1033–1037. https://doi.org/10.1016/j.
bbi.2007.04.009.
Raison, C.L., Knight, J.M., Pariante, C., 2018. Interleukin (IL)-6: a good kid hanging out
with bad friends (and why sauna is good for health). Brain. Behav. Immun. 73, 1–2.
https://doi.org/10.1016/j.bbi.2018.06.008.
Rivera, J., Alegre, C., Ballina, F.J., Carbonell, J., Carmona, L., Castel, B., Collado, A.,
Esteve, J.J., Martínez, F.G., Tornero, J., Vallejo, M.A., Vidal, J., 2006. Documento de
consenso de la Sociedad Espa˜
nola de Reumatología sobre la bromialgia. Reumatol.
Clin. 2 (Suppl 1), S55–S66. https://doi.org/10.1016/S1699-258X(06)73084-4.
Rodriguez-Pinto, I., Agmon-Levin, N., Howard, A., Shoenfeld, Y., 2014. Fibromyalgia and
cytokines. Immunol. Lett. 161, 200–203. https://doi.org/10.1016/j.
imlet.2014.01.009.
Rogerson, M., Gladwell, V.F., Gallagher, D.J., Barton, J.L., 2016. Inuences of green
outdoors versus indoors environmental settings on psychological and social
outcomes of controlled exercise. Int. J. Environ. Res. Public Health 13, 363. https://
doi.org/10.3390/ijerph13040363.
Sanabria-Mazo, J.P., Montero-Marin, J., Feliu-Soler, A., Gasi´
on, V., Navarro-Gil, M.,
Morillo-Sarto, H., Colomer-Carbonell, A., Borr`
as, X., Tops, M., Luciano, J.V., García-
Campayo, J., 2020. Mindfulness-based program plus amygdala and insula retraining
(MAIR) for the treatment of women with bromyalgia: a pilot randomized controlled
trial. J. Clin. Med. 9, 3246. https://doi.org/10.3390/jcm9103246.
Sanada, K., Díez, M.A., Valero, M.S., P´
erez-Yus, M.C., Demarzo, M.M., García-Toro, M.,
García-Campayo, J., 2015. Effects of non-pharmacological interventions on
inammatory biomarker expression in patients with bromyalgia: a systematic
review. Arthritis Res. Ther. 17, 272. https://doi.org/10.1186/s13075-015-0789-9.
Sawaddiruk, P., Paiboonworachat, S., Chattipakorn, N., Chattipakorn, S.C., 2017.
Alterations of brain activity in bromyalgia patients. J. Clin. Neurosci. 38, 13–22.
https://doi.org/10.1016/j.jocn.2016.12.014.
Schulz, K.F., Altman, D.G., Moher, D., CONSORT Group., 2010. CONSORT 2010
statement: updated guidelines for reporting parallel group randomized trials. BMJ.
340, c332. https://doi.org/10.1136/bmj.c332.
Schwamm, L.H., Erskine, A., Licurse, A., 2020. A digital embrace to blunt the curve of
COVID19 pandemic. NPJ Digit Med. 3, 64. https://doi.org/10.1038/s41746-020-
0279-6.
Serrano-Atero, M.S., Caballero, J., Ca˜
nas, A., García-Saura, P.L., Serrano-´
Alvarez, C.,
Prieto, J., 2002. Pain assessment (I). Rev Soc Esp Dolor. 9, 94–108.
Serrat, M., Almirall, M., Must´
e, M., Sanabria-Mazo, J.P., Feliu-Soler, A., M´
endez-
Ulrich, J.L., Luciano, J.V., Sanz, A., 2020. Effectiveness of a multicomponent
treatment for bromyalgia based on pain neuroscience education, exercise therapy,
psychological support, and nature exposure (NAT-FM): a pragmatic randomized
controlled trial. J. Clin. Med. 9, 3348. https://doi.org/10.3390/jcm9103348.
Serrat, M., Coll-Oma˜
na, M., Albajes, K., Sol´
e, S., Almirall, M., Luciano, J.V., Feliu-
Soler, A., 2021a. Efcacy of the FIBROWALK multicomponent program moved to a
virtual setting for patients with bromyalgia during the COVID-19 pandemic: a
proof-of-concept RCT performed alongside the state of alarm in Spain. J. Environ.
Res. Public Health. 18, 10300. https://doi.org/10.3390/ijerph181910300.
S. Ferr´
es et al. Brain Behavior and Immunity 125 (2025) 184–197
196
Serrat, M., Sanabria-Mazo, J.P., Almirall, M., Must´
e, M., Feliu-Soler, A., M´
endez-
Ulrich, J.L., Sanz, A., Luciano, J.V., 2021b. Effectiveness of a multicomponent
treatment based on pain neuroscience education, therapeutic exercise, cognitive
behavioral therapy, and mindfulness in patients with bromyalgia (FIBROWALK
Study): a randomized controlled trial. Phys. Ther. 101, pzab200. https://doi.org/
10.1093/ptj/pzab200.
Serrat, M., Albajes, K., Navarrete, J., Almirall, M., Lluch Girb´
es, E., Neblett, R.,
Luciano, J.V., Moix, J., Feliu-Soler, A., 2022a. Effectiveness of two video-based
multicomponent treatments for bromyalgia: the added value of cognitive
restructuring and mindfulness in a three-arm randomized controlled trial. Behav.
Res. Ther. 158, 104188. https://doi.org/10.1016/j.brat.2022.104188.
Serrat, M., Ferr´
es, S., Auer, W., Almirall, M., Lluch, E., D’Amico, F., Maes, M., Lorente, S.,
Navarrete, J., Montero-Marín, J., Neblett, R., Nijs, J., Borr`
as, X., Luciano, J.V., Feliu-
Soler, A., 2022b. Effectiveness, cost-utility and physiological underpinnings of the
FIBROWALK multicomponent therapy in online and outdoor format in individuals
with bromyalgia: study protocol of a randomized, controlled trial (On&Out study).
Front. Physiol. 13, 1046613. https://doi.org/10.3389/fphys.2022.1046613.
Sharpe, L., Jones, E., Ashton-James, C.E., Nicholas, M.K., Refshauge, K., 2020. Necessary
components of psychological treatment in pain management programs: a Delphi
study. Eur. J. Pain 24, 1160–1168. https://doi.org/10.1002/ejp.1561.
Shields, G.S., Spahr, C.M., Slavich, G.M., 2020. Psychosocial interventions and immune
system function: a systematic review and meta-analysis of randomized clinical trials.
JAMA Psychiat. 77, 1031–1043. https://doi.org/10.1001/
jamapsychiatry.2020.0431.
Sluka, K.A., Clauw, D.J., 2016. Neurobiology of bromyalgia and chronic widespread
pain. Neuroscience 338, 114–129. https://doi.org/10.1016/j.
neuroscience.2016.06.006.
Stanhope, J., Breed, M.F., Weinstein, P., 2020. Exposure to greenspaces could reduce the
high global burden of pain. Environ. Res. 187, 109641. https://doi.org/10.1016/j.
envres.2020.109641.
Thase, M.E., 2014. Using biomarkers to predict treatment response in major depressive
disorder: evidence from past and present studies. Dialogues Clin. Neurosci. 16,
539–544. https://doi.org/10.31887/DCNS.2014.16.4/mthase.
Thieme, K., Mathys, M., Turk, D.C., 2017. Evidenced-based guidelines on the treatment
of bromyalgia patients: are they consistent and if not, why not? have effective
psychological treatments been overlooked? J. Pain 18, 747–756. https://doi.org/
10.1016/j.jpain.2016.12.006.
Thompson Coon, J., Boddy, K., Stein, K., Whear, R., Barton, J., Depledge, M.H., 2011.
Does participating in physical activity in outdoor natural environments have a
greater effect on physical and mental wellbeing than physical activity indoors? a
systematic review. Environ. Sci. Technol. 45, 1761–1772. https://doi.org/10.1021/
es102947t.
Tkachuk, G.A., Harris, C.A., 2012. Psychometric properties of the tampa scale for
kinesiophobia-11 (TSK-11). J. Pain 13, 970–977. https://doi.org/10.1016/j.
jpain.2012.07.001.
Trøstrup, C.H., Christiansen, A.B., Stølen, K.S., Nielsen, P.K., Stelter, R., 2019. The effect
of nature exposure on the mental health of patients: a systematic review. Qual. Life
Res. 28, 1695–1703. https://doi.org/10.1007/s11136-019-02125-9.
Twisk, J., de Boer, M., de Vente, W., Heymans, M., 2013. Multiple imputation of missing
values was not necessary before performing a longitudinal mixed-model analysis.
J. Clin. Epidemiol. 66, 1022–1028. https://doi.org/10.1016/j.jclinepi.2013.03.017.
White, V., Linardon, J., Stone, J.E., Holmes-Truscott, E., Olive, L., Mikocka-Walus, A.,
Hendrieckx, C., Evans, S., Speight, J., 2022. Online psychological interventions to
reduce symptoms of depression, anxiety, and general distress in those with chronic
health conditions: a systematic review and meta-analysis of randomized controlled
trials. Psychol. Med. 52, 548–573. https://doi.org/10.1017/S0033291720002251.
Worrel, L.M., Krahn, L.E., Sletten, C.D., Pond, G.R., 2001. Treating bromyalgia with a
brief interdisciplinary program: initial outcomes and predictors of response. Mayo
Clin. Proc. 76, 384–390. https://doi.org/10.4065/76.4.384.
Xiong, H.Y., Hendrix, J., Schabrun, S., Wyns, A., Campenhout, J.V., Nijs, J., Polli, A.,
2024. The role of the brain-derived neurotrophic factor in chronic pain: links to
central