TYPE Original Research PUBLISHED 16 March 2023 DOI 10.3389/fnins.2023.1148176 Brain system segregation and OPEN ACCESS pain catastrophizing in chronic EDITED BY Hongxing Zhang, Xuzhou Medical University, China pain progression REVIEWED BY Junqin Ma, Selma Delgado-Gallén1,2,3*, MD Soler1,2,3 , Southern Medical University, China María Cabello-Toscano1,4,5 , Kilian Abellaneda-Pérez1,2,3 , Hongrui Zhan, The Fifth Affiliated Hospital of Sun Yat-sen Javier Solana-Sánchez1,2,3 , Goretti España-Irla1,2,3 , University, China Alba Roca-Ventura1,4,5 , David Bartrés-Faz1,4,5 , *CORRESPONDENCE Selma Delgado-Gallén Josep M. Tormos1,6 , Alvaro Pascual-Leone1,7,8 and selmadelgado@guttmann.com Gabriele Cattaneo1,2,3 SPECIALTY SECTION 1 This article was submitted to Institut Guttmann, Institut Universitari de Neurorehabilitació adscrit a la UAB, Barcelona, Spain, 2 Perception Science, Departament de Medicina, Facultat de Medicina, Universitat Autónoma de Barcelona, Bellaterra, Spain, 3 a section of the journal Fundació Institut d’Investigació en Ciències de la Salut Germans Trias i Pujol, Barcelona, Spain, 4 Institut Frontiers in Neuroscience d’Investigacions Biomèdiques August Pi i Sunyer (IDIBAPS), Barcelona, Spain, 5 Departament de Medicina, Facultat de Medicina i Ciéncies de la Salut, Universitat de Barcelona, Barcelona, Spain, 6 Centro RECEIVED 19 January 2023 de Investigación Traslacional San Alberto Magno, Facultad de Medicina y Ciencias de la Salud, ACCEPTED 28 February 2023 Universidad Católica de Valencia San Vicente Mártir, Valencia, Spain, 7 Hinda and Arthur Marcus Institute PUBLISHED 16 March 2023 for Aging Research and Center for Memory Health, Hebrew SeniorLife, Boston, MA, United States, 8 CITATION Department of Neurology, Harvard Medical School, Boston, MA, United States Delgado-Gallén S, Soler MD, Cabello-Toscano M, Abellaneda-Pérez K, Solana-Sánchez J, España-Irla G, Roca-Ventura A, Bartrés-Faz D, Tormos JM, Pascual-Leone A and Cattaneo G (2023) Brain Pain processing involves emotional and cognitive factors that can modify system segregation and pain catastrophizing pain perception. Increasing evidence suggests that pain catastrophizing (PC) in chronic pain progression. Front. Neurosci. 17:1148176. is implicated, through pain-related self-thoughts, in the maladaptive plastic doi: 10.3389/fnins.2023.1148176 changes related to the maintenance of chronic pain (CP). Functional magnetic COPYRIGHT resonance imaging (fMRI) studies have shown an association between CP and © 2023 Delgado-Gallén, Soler, Cabello-Toscano, Abellaneda-Pérez, two main networks: default mode (DMN) and dorsoattentional (DAN). Brain Solana-Sánchez, España-Irla, Roca-Ventura, system segregation degree (SyS), an fMRI framework used to quantify the extent Bartrés-Faz, Tormos, Pascual-Leone and Cattaneo. This is an open-access article to which functional networks are segregated from each other, is associated distributed under the terms of the Creative with cognitive abilities in both healthy individuals and neurological patients. Commons Attribution License (CC BY). The We hypothesized that individuals suffering from CP would show worst health- use, distribution or reproduction in other forums is permitted, provided the original related status compared to healthy individuals and that, within CP individuals, author(s) and the copyright owner(s) are longitudinal changes in pain experience (pain intensity and affective interference), credited and that the original publication in this journal is cited, in accordance with accepted could be predicted by SyS and PC subdomains (rumination, magnification, and academic practice. No use, distribution or helplessness). To assess the longitudinal progression of CP, two pain surveys were reproduction is permitted which does not comply with these terms. taken before and after an in-person assessment (physical evaluation and fMRI). We first compared the sociodemographic, health-related, and SyS data in the whole sample (no pain and pain groups). Secondly, we ran linear regression and a moderation model only in the pain group, to see the predictive and moderator values of PC and SyS in pain progression. From our sample of 347 individuals (mean age = 53.84, 55.2% women), 133 responded to having CP, and 214 denied having CP. When comparing groups, results showed significant differences in health-related questionnaires, but no differences in SyS. Within the pain group, helplessness (β = 0.325; p = 0.003), higher DMN (β = 0.193; p = 0.037), and lower DAN segregation (β = 0.215; p = 0.014) were strongly associated with a worsening in pain experience over time. Moreover, helplessness moderated the association between DMN segregation and pain experience progression (p = 0.003). Our findings indicate that the efficient functioning of these networks Frontiers in Neuroscience 01 frontiersin.org Delgado-Gallén et al. 10.3389/fnins.2023.1148176 and catastrophizing could be used as predictors of pain progression, bringing new light to the influence of the interplay between psychological aspects and brain networks. Consequently, approaches focusing on these factors could minimize the impact on daily life activities. KEYWORDS chronic pain, functional magnetic resonance imaging, resilience, system segregation, pain catastrophizing, psychological distress 2019; Spisak et al., 2019; You et al., 2021; De Ridder et al., 2022; 1. Introduction Solé-Padullés et al., 2022). Crucially it has been suggested that two specific main large-scale brain networks are related to altered pain Chronic pain (CP) is an unpleasant sensory and emotional processing and pain chronification: the dorsoattentional network experience associated with negative cognitive and emotional (DAN), and the default mode network (DMN) (Spreng et al., 2013; aspects like the feeling of unpleasantness, pain catastrophizing Baliki et al., 2014; Becerra et al., 2014; Kucyi et al., 2014; Kilpatrick (PC), and decreased physical and psychological functioning across et al., 2015; Seidler et al., 2015; Franzmeier et al., 2017; Kim et al., the lifespan (Dahlhamer, 2018). This prolonged pain experience 2019; van Ettinger-Veenstra et al., 2019; Jones et al., 2020; De encompasses pain intensity (i.e., how much a patient is in pain) and Ridder et al., 2022). pain affection aspects, the degree of emotional arousal caused by DAN network is prominently involved in goal-directed the sensory experience of pain (Haefeli and Elfering, 2006). attention and top-down selection of stimuli and responses, These aspects influence pain perception and modulation, and interacting dynamically with the salience network (SN) and vice-versa (Wiech, 2016, 2018; Opdebeeck et al., 2018; Wiech and control executive network. Concretely, DAN is responsible for the Shriver, 2018; Delgado-Gallén et al., 2021), resulting in large inter- maintenance of spatial priority maps for covert spatial attention, individual differences in terms of pain assessment and treatment. saccade planning, and visual working memory (Vossel et al., 2014). Thus, in the past decades, PC and functional brain connectivity In CP patients it has been found that the functional connectivity (FC) has been shown to play an important role in understanding of this network, especially concerning its connectivity with other pain experience, its chronification, and treatment response. large-scale networks, is consistently altered (Coppola et al., 2019; Pain catastrophizing, intended as the tendency to magnify and Mao et al., 2022), and could normalize after pain therapy (Yoshino ruminate about pain, boasts attentional biases to threatening et al., 2018). aspects of painful experience (Gilliam et al., 2017) and, On the other hand, the DMN, which controls self- consequently, impacts self-reported pain intensity and pain- representational processing, is normally deactivated during related psychological aspects (Jensen et al., 2017; Suso-Ribera task or stimulus exposure, but not in CP patients (Legrain et al., et al., 2017; Häggman-Henrikson et al., 2021). Therefore, the 2009; Seminowicz et al., 2011; Kucyi et al., 2014; De Ridder combination of PC and unpleasantness can lead to suffering, anger, et al., 2022), exhibiting abnormal DMN resting-state functional fear, frustration, or anxiety, having an important role in adjustment connectivity (rs-FC). Alteration in the connectivity of this network to CP (Ziadni et al., 2018) and pain management (Gilliam et al., has been coupled to several pain types (e.g., low back pain, complex 2017). regional pain syndrome (CRPS), chronic widespread pain, or In this line, it has been suggested that the relationship between osteoarthritic pain), potentially explaining why pain becomes an pain and depressed mood is mediated by PC (Dong et al., 2020). integral part of the self. Finally, DMN has been also consistently In addition, studies exploring the differential relationships among associated with pain anticipation, pain intensity, and PC (Napadow individual components of PC (helplessness, magnification, and et al., 2010; Ter Minassian et al., 2012; Loggia et al., 2013; Kucyi rumination) and pain outcomes (Müller, 2011; Craner et al., 2016; et al., 2014; De Ridder et al., 2022). Stensland, 2021), indicated that catastrophic thoughts and behavior In this study we aim to use a novel approach in the are as well linked with pain co-morbidities as sleep (Abeler et al., study of the relationship between CP, PC, and brain networks 2020), mental health (Dong et al., 2020) and physical activity. functional connectivity, using rs-fMRI combined with graph theory Furthermore, the predictive value of PC in pain evolution has been methods (i.e., the topological organization of brain networks) found in several studies. For instance, high levels of PC predicted a (Wang et al., 2010; Ewers et al., 2021). Concretely we will decrease in physical activity due to sedentary behavior (Zhaoyang study networks system segregation (SyS) (Wig, 2017), a paradigm et al., 2020), or higher levels of acute or persistent pain after surgery based on the idea that effective network functioning is supported for knee osteoarthritis (Burns et al., 2015). by maintaining subnetworks’ segregation while simultaneously Also, the neural correlates of these phenomena have been allowing integration between them (Wig, 2017), resulting in a explored in multiple studies. Altered brain connectivity in pain brain more adaptable to task demands. Brain networks’ segregation disorders in task-oriented and resting-state functional magnetic degree has been previously associated with cognitive abilities in resonance imaging (rs-fMRI) has been shown altered in the healthy adults and patients affected by neurologic diseases (van den presence of acute and CP (Baliki et al., 2014; Kucyi et al., 2014; Heuvel and Sporns, 2013; Malagurski et al., 2020; Ewers et al., 2021; Yu et al., 2014; Hemington et al., 2016; Pfannmöller and Lotze, Riedel et al., 2021). Frontiers in Neuroscience 02 frontiersin.org Delgado-Gallén et al. 10.3389/fnins.2023.1148176 From our knowledge, this is the first study that explores how 2.2. Assessments SyS, the ratio of rs-FC within brain networks and their connection with the rest of the cortex, could be related to the presence of pain, 2.2.1. Pain: Clinical symptoms and the potential role of SyS and pain-related psychological factors Participants were screened online for CP, which was considered in longitudinal changes in the pain experience. the pain that persists or recurs for > 3 months according to the International Association for the Study of Pain (IASP) criteria (Treede et al., 2015). 2. Materials and methods Only those participants who answered to have recurring or persisting pain underwent the posterior pain assessment explained 2.1. Participants and study design below. Thus, the pain group in this study was formed by participants who answered to have pain during at least 3 months This study was performed in the framework of the Barcelona in both questionnaires (T1 and T2), while participants who had Brain Health Initiative (BBHI) (Cattaneo et al., 2018, 2020), an not CP in either of the questionnaires were included in the no pain ongoing prospective longitudinal study that started in 2017 to group. Non-recurrent or non-persistent pain (i.e., acute pain) was identify lifestyle factors and biological mechanisms underlying not assessed in this study, neither in healthy nor in CP subjects. good brain health in middle-aged adults (40–65 years). Between The Brief Pain Inventory-Short Form (BPI-SF) (Keller et al., 2018 and 2021 participants underwent an online self-assessment 2004) was used to assess the intensity and severity of pain and (through the BBHI web-based platform) of sociodemographic pain onset. Validation studies among CP patients and the published characteristics, mental health (MH), quality of life (QoL), and Spanish translation demonstrated good psychometric properties self-perceived cognitive concerns (see Figure 1 for more details). (Badia et al., 2003). Pain intensity was estimated considering the mean of pain Between 2019 and 2020 they answered the baseline pain intensity in the last week, considering that asking for a short past questionnaire (T1), and from 2020 to 2021 the second one (T2). period (i.e., 1 week) is more reliable than asking for “current” pain Parallelly, between May 2018 and February 2021 (T10 ), (Haefeli and Elfering, 2006). participants performed an in-person assessment, including a Pain interference was estimated through seven domains divided medical exam and MRI. The same in-person evaluation protocol into two subdimensions (with the arithmetic mean): affective was applied to all participants. (relations with others, enjoyment of life, sleep, and mood) and Online questionnaires and in-person assessments were not activity (walking, general activity, and work), according to the BPI paralleled in time and were not equal for all participants. The time user guide (Cleeland and Ryan, 1991). Although sleep is seen as a gap between T1 and T10 was approximately 4 months, and between third domain in some studies, we used it in the affective domain T10 and T2, 14 months. However, the minimum distance between (Miettinen et al., 2019). questionnaires, and between the fMRI scan and T2 was fixed at Questions about the number of pain sites (one, two, three, or 3 months to leave room for a reliable pain progression and assure four or more pain sites), and pain medication (non-steroidal anti- that the criteria for the diagnosis of CP (3 months) were satisfied in inflammatory, anti-migraines. . .) were added to the questionnaire. each questionnaire independently from the other. The first and second assessments were conducted during Participants with CP were excluded if (1) pain was due to 2019–2020 (T1) and 2020–2021 (T2), respectively, (approximately cancer, fracture, infection, or diagnosis of neurologic or psychiatric 10 months between both, see Figure 1). disease, (2) made active substance abuse disorder in the past 2 years, (3) uses of prescription opioids exceeding 60 mg morphine 2.2.2. Pain experience over time: Intensity and equivalents per day, (4) were not suitable for MRI scan and (5) they affective interference had excessive head motion during the MRI (at least in the 50% of To calculate the longitudinal progression of the pain the scans movements were above a 0.5 mm threshold). experience, we calculated the mean between pain intensity From our sample, 450 participants completed online and in- and pain affective interference at each time point: person assessments (see Figure 1): 133 consistently answered having CP (see below for a CP definition) that persisted for at Pain experience 1 = least 3 months in both questionnaires, 216 without CP in any Pain intentistyT1 − Pain affective interferenceT1 time-point (where 2 participants were excluded for excessive head 2 motion during MRI), 39 participants that developed CP (that is, they first answered not having CP, and in the second questionnaire they did), and 62 that recovered from a CP (i.e., they started with Pain experience 2 = CP and then they reported not having pain anymore). Considering Pain intentistyT2 − Pain affective interferenceT2 our principal aim to study pain chronification and its maintenance, the present work included 347 volunteers (133 with CP and 214 2 without CP). Then we subtracted the numerical value of the first assessment The protocol was approved by the “Comité d’Ètica i from the second assessment, resulting in a worsening if the final Investigació Clínica de la Unió Catalana d’Hospitals” and was score was positive or an amelioration if it was negative. carried out following the Declaration of Helsinki (World Medical Pain experience longitudinal = Association, 2013). Written informed consent was obtained from all participants before inclusion in the study. Pain experience 2−Pain experience 1 Frontiers in Neuroscience 03 frontiersin.org Delgado-Gallén et al. 10.3389/fnins.2023.1148176 FIGURE 1 Flowchart of the selection and distribution of patients in this study. Chart illustrating how participants were asked to fill questionnaires (T1 and T2) and undertake an MRI (T10 ), and then formed the pain group (YES-YES) and the no pain group (NO-NO). Those participants who did not maintain pain on both questionnaires were not included on this study (YES-NO, NO-YES). Mean time between first questionnaire (T1) and the MRI (T10 ) was approximately 4 months, and between the MRI and the second questionnaire (T2), 14 months. *Two participants were excluded from the no pain group for having excessive motion (at least the 50% of the between scans movements were above a 0.5 mm threshold). Frontiers in Neuroscience 04 frontiersin.org Delgado-Gallén et al. 10.3389/fnins.2023.1148176 2.2.3. Pain catastrophizing inversion time = 1,000 ms, flip angle = 8◦ , a field of view Pain catastrophizing can be understood as an expansion of (FOV) = 256 mm and 0.8 mm isotropic voxel]. Additionally, a maladaptive cognitive response during actual or perceived painful high-resolution 3D SPC T2 weighted acquisition was undertaken stimuli, comprising negative cognitive and emotional processes. (TR = 3,200 ms, TE = 563 ms, flip angle = 120◦ , 0.8 mm The PC scale is a multidimensional construct that encompasses isotropic voxel, FOV = 256 mm). In the same session, they also elements of rumination, magnification, and helplessness. It can be underwent rs-fMRI multiband (anterior-posterior phase-encoding; computed by summing responses to all 13 items, and uses a 5-point acceleration factor = 8) interleaved acquisitions (T2∗ weighted EPI scale, ranging from 0 (not at all) to 4 (all the time). The total scale scans, TR = 800 ms, TE = 37 ms, 750 volumes, 72 slices, slice score ranges from 0 to 52, with higher scores representing greater thickness = 2 mm, FOV = 208 mm). All the MRI images were catastrophic thinking. Participants are asked to indicate the degree examined by a senior neuroradiologist (NB) to detect any clinically to which they experienced each of the 13 thoughts or feelings when significant pathology (none found). Then, all the acquisitions were experiencing pain. This scale has been demonstrated to have good visually inspected before analysis (MC-T and LM-P.) to ensure that psychometric proprieties (Sullivan et al., 1995; García Campayo they did not contain MRI artifacts or excessive motion. et al., 2008; Olmedilla Zafra et al., 2013) and it has been validated in the Spanish language. 2.3.2. MRI preprocessing The rs-fMRI preprocessing pipeline comprised spatial 2.2.4. Health-related questionnaires standardization and nuisance correction by making use of We employed the ultra-brief self-reported Patient Health functions from FMRIB Software Library (FSL; version 5.0.11)1 , Questionnaire-4 (PHQ-4) to assess MH. The PHQ-4 had four items FreeSurfer (version 6.0)2 and Statistical Parametric Mapping asking about mood disorder symptoms (two items for depression (SPM12).3 To start with, the first 10 scans were removed to and the other two items for anxiety) in the past 2 weeks. All ensure magnetization equilibrium. After that, all images were field items were rated on a four-point scale ranging from 0 (not at all) inhomogeneity corrected (FSL topup tool), all scans realigned to a to 3 (nearly every day). The published Spanish translation also reference image (FSL MCFLIRT), and then standardized into native demonstrated good psychometric properties in a validation study T1-weighted space (SPM Coregister). Finally, normalization (SPM (Kocalevent et al., 2014). Normalize) of all fMRI images to Montreal Neuroscience Institute To assess the QoL, we used WHO-QoL-AGE, a 13-item (MNI152) standard space was performed to ensure among-subjects questionnaire scored on a 5-point Likert scale. This questionnaire comparability. As for nuisance correction, different components attempts to assess satisfaction with one’s life, living place, were defined and manually removed from the rs-fMRI images general health (i.e., hearing, vision. . .), daily activities, personal by the “fsl_regfilt” tool implemented in FSL. These components relationships, goal achievements, or economic status (The Whoqol correspond to (1) motion regressors of rotation, translation, Group, 1998). The published Spanish translation has good and their derivatives, as estimated during scans’ realignment, psychometric data in a validation study (Lucas-Carrasco, 2012). (2) a drift estimated by a discrete cosine transform (DCT) as a We used PROMIS Cognitive Abilities and Cognitive Concerns R low-pass frequency filter (< 0.01), and (3) signals from white scales to assess cognitive troubles. This scale consists of 12- matter (WM) and cerebrospinal fluid (CSF). To extract these, item, extracted from the PROMIS item bank, measuring self- CSF and WM masks were obtained from automatic subcortical reported cognitive deficits in memory, working memory, attention, segmentation of brain volume, based on the existence of an atlas processing speed, or cognitive flexibility (Fieo et al., 2016). Each containing probabilistic information on the location of structures item asked participants to answer “within the last 7 days” using five (Fischl et al., 2002). This step was part of the FreeSurfer “recon- response options. all” processing stream, which was run with default parameters, except for the addition of the T2 flag for the improvement of pial surfaces reconstruction. Both T1-w and T2-w images were used for processing anatomical information. 2.3. MRI acquisition and analysis As head movement may affect rs-fMRI results (Van Dijk et al., 2012; Power et al., 2013, 2014, 2015), the in-scanner head motion 2.3.1. MRI acquisition parameters was considered. In this study, the framewise displacement (FWD) Magnetic resonance imaging data were acquired in a 3T mean was calculated for every subject. FWD was computed as in Siemens scanner (MAGNETOM Prisma) (Siemens Healthcare Power et al. (2012), using the vectors of rotation and translation GmbH, Erlangen, Germany) with a 32-channel head coil, at the estimated during scans’ realignment as part of the preprocessing Unitat d’Imatge per Ressonància Magnètica IDIBAPS (Institut pipeline (Power et al., 2012). d’Investigacions Biomèdiques August Pi i Sunyer) at Hospital Clínic de Barcelona, Barcelona. MRI session included accelerated 2.3.3. Functional magnetic resonance imaging: multi-band sequences adapted from the Human Connectome System segregation Project and provided by the Center of Magnetic Resonance A node-based approach was adopted to quantify subject rs- Research at the University of Minnesota. For all participants, FC and SyS of seven resting-state networks (RSN) as defined a high-resolution T1-weighted structural image was obtained with a magnetization-prepared rapid acquisition gradient-echo (MPRAGE) three-dimensional protocol, and a total of 208 1 https://fsl.fmrib.ox.ac.uk/fsl/fslwiki/ contiguous axial slices were obtained in ascending fashion 2 https://surfer.nmr.mgh.harvard.edu [repetition time (TR) = 2,400 ms, echo time (TE) = 2.22 ms, 3 https://www.fil.ion.ucl.ac.uk/spm/ Frontiers in Neuroscience 05 frontiersin.org Delgado-Gallén et al. 10.3389/fnins.2023.1148176 in the Yeo atlas (Thomas Yeo et al., 2011). These functional 2.4.4. Regression models within the chronic pain connectivity measures have been previously used to study brain group networks implicated in pain progression (Kastrati et al., 2022; Lee To explore which variables affected pain experience over time, et al., 2022). To increase Yeo-atlas spatial resolution and precision we ran a first multiple regression model with pain experience as the at rs-FC computation, the 100 nodes, and 7 networks Schaeffer- primary outcome, and sociodemographic data (age, biological sex, Yeo atlas was used (Figure 2; Schaefer et al., 2018).4 Per each of and BMI), pain baseline ratings (pain-related medication, intensity, the 100 regions of interest (ROIs), a BOLD signal was extracted total interference, duration, and the number of pain sites), PC and averaged across all voxels falling within an ROI. Then, ROI- scale total score and health-related questionnaires (MH, QoL, and to-ROI rs-FCs were computed as Pearson-Moment correlations, cognitive concerns) as regressors. and subsequently, Fisher-z transformed. Negative values were set Then, we ran different multiple regression models using pain to zero and autocorrelations were not considered in the further experience progression (Pain experience 2- Pain experience 1) as computation. the dependent variable and sociodemographic (age and biological System segregation (SyS) is a graph theory metric to quantify sex), PC subscales, pain characteristics at baseline (intensity, the extent to which major functional networks are segregated from affective interference, and activity interference), and SyS. We also each other (see Figure 3). As expressed in added covariable head motion during the MRI, as it has an important influence on intrinsic functional connectivity and its Wnet − Bnet interpretation (Van Dijk et al., 2012), in the model and the time SySnet = , Wnet between the first questionnaire (T1) and MRI (T10 ). Finally, we added a multicollinearity diagnostic to ensure the System segregation (SyS) captures the balance between within- interdependence of regressors using the variance inflation factor network (Wnet ) and between-networks (Bnet ) rs-FC. Within- coefficient. Variables in this step were selected from previous network rs-FC was computed as the average rs-FC connecting all analyses depending on if they had achieved significant effects the nodes within the same network. Between-network rs-FC was (like pain intensity, interference, and catastrophizing) or if they computed as the average rs-FC connecting nodes of a particular could alter SyS results (age, biological sex, distance between first network to nodes from the rest of the cortex. questionnaire and fMRI, and head motion during fMRI). To explore the interaction between catastrophism’s helplessness and brain networks related to pain progression, we run moderation 2.4. Data analysis models through SPSS Process , and we look for the Johnson- R Neyman interval. Finally, to explore the possible effect of pain 2.4.1. Comparisons between no pain and pain experience in interaction with catastrophism’s helplessness on groups brain connectivity we repeated the models using brain SyS as We first analyzed group differences (no pain and CP) in the dependent variable and pain experience as the independent sociodemographic data [age, biological sex, and body mass variable. index (BMI)], health-related questionnaire scores (MH, cognitive complaints, and QoL), and head movements during the scan using Welch t-tests and chi-square statistics. In addition, we also ran a 2.4.5. Statistical analyses Statistical analyses were performed using SPSS version 20.0 multivariate analysis between these groups on the segregation of (Statistical Package for Social Sciences, Chicago, IL, USA). networks, correcting for the delay between the first questionnaire and MRI, head movement during MRI, age, and biological sex. Regression models’ graphics were created through R v.3.6.3 (R Foundation for Statistical Computing, Vienna, Austria). Brain 2.4.2. Correlations between pain variables, nodes and segregation graphic representation were created with the Surf Ice tool (version 6-October-2021; v1.0.20211006), which is an psychological aspects, and system segregation OpenGL Shading Language surface rendering source code. We ran bivariate correlations to explore the association between psychological variables (MH, QoL, cognitive concerns), PC subscales, and pain variables at baseline (intensity, interference, number of pain sites, duration, and medication intake). Besides, 3. Results we ran partial correlations between SyS and pain factors (intensity, interference, and the number of pain sites), correcting for age, 3.1. Differences between no pain and gender, and head motion. chronic pain group 2.4.3. Description of longitudinal pain progression When we compared the no pain and CP group for We explored the differences between pain characteristics sociodemographic data and health-related questionnaires, we (intensity, affective and activity interference, number of pain sites, found significant differences in age (F = 6,271; p = 0.013), gender pain duration, and pain-related medication) in both time points (x2 = 8.396; p = 0.003), BMI (F = 9,352; p = 0.002), MH (F = 24,772; using t-tests and chi-square statistics. p < 0.001), cognitive complaints (F = 31,945; p < 0.001), QoL (F = 40,392; p < 0.001) and head movement during the scan 4 https://github.com/ThomasYeoLab/CBIG/tree/master/stable_projects/ (F = 4,652; p = 0.032). CP group was composed predominantly brain_parcellation/Schaefer2018_LocalGlobal of women (63.2%) and older people when compared with the no Frontiers in Neuroscience 06 frontiersin.org Delgado-Gallén et al. 10.3389/fnins.2023.1148176 FIGURE 2 Nodes of the seven networks. Nodes of each of the studied networks based on the Schaffer-Yeo atlas of 100 regions of interest and 7 networks. FIGURE 3 Nodes of the seven networks. A node-based approach was adopted to quantify subject rs-FC and SyS of seven resting-state networks (RSN) as defined in the Yeo atlas (Thomas Yeo et al., 2011). Per each of the 100 regions of interest (ROIs), a BOLD signal was extracted and averaged across all voxels falling within an ROI. Then, ROI-to-ROI rs-FCs were computed as Pearson-moment correlations, and subsequently, Fisher-z transformed. Negative values were set to zero and autocorrelations were not considered in the further computation. SyS captures the balance between within-network (Wnet ) and between-networks (Bnet ) rs-FC. Within-network rs-FC was computed as the average rs-FC connecting all the nodes within the same network. Between-network rs-FC was computed as the average rs-FC connecting nodes of a particular network to nodes from the rest of the cortex. pain group (mean = 54.81; SD = 7.04) (detailed information is in and interference (r = −0.521, p < 0.001), cognitive complaints Table 1). (r = 0.398, p < .001), MH (r = −0.533, p < 0.001), rumination Regarding SyS, the multivariate analysis (corrected by head (r = −0.250, p = 0.004), magnification (r = −0.442, p < 0.001) and movement, the time between the first questionnaire, biological sex, helplessness (r = −0.478, p < 0.001). No correlation was present and age; see Table 1) showed no differences between groups in between QoL and pain duration (r = −0.011, p = 0.900) or pain DMN (F = 0.046; p = 0.831), somatomotor network (F = 0.026; medication (r = −0.087, p = 0.321). p = 0.871), control network (F = 0.772; p = 0.380), DAN (F = 0.390; For MH, we found significant correlations with multiple pain p = 0.533), SN (F = 2.466; p = 0.117), limbic network (F = 0.050; sites (r = 0.184, p = 0.034), pain medication (r = 0.187, p = 0.031), p = 0.822) and visual network (F = 0.000; p = 0.999) (Table 1). pain intensity (r = 0.213, p = 0.014), pain interference (r = 0.371, p < 0.001), QoL, cognitive concerns (r = −0.423, p < 0.001), rumination (r = 0.288, p = 0.001), magnification (r = 0.471, 3.2. Correlations in psychological aspects p < 0.001) and helplessness (r = 0.422, p < 0.001). No correlation and pain variables was found with pain duration. Finally, cognitive complaints were positively correlated with Quality of life significantly correlated with multiple pain sites multiple pain sites (r = −0.182, p = 0.038), MH, QoL, magnification (r = −0.243 p = 0.005), pain intensity (r = −0.282, p = 0.001) (r = −0.222, p = 0.011), and helplessness (r = −0.191, p = 0.029). Frontiers in Neuroscience 07 frontiersin.org Delgado-Gallén et al. 10.3389/fnins.2023.1148176 Concerning SyS and pain factors (intensity, interference, and any kind of treatment at that moment. Finally, 32.3% of our the number of pain sites) associations, we only found pain sample treated their pain with other techniques, like physical or interference and SyS of DMN (r = 0.186, p = 0.034) and SN psychological therapies. The mean catastrophizing total score was (r = 0.215, p = 0.014), and a tendency of an association with DAN 16.44 (SD = 8.65), helplessness 6.41 (SD = 4.19), rumination 6.44 (r = 0.161, p = 0.068). (SD = 3.37), and magnification 3.60 (SD = 2.26). When comparing two-time points (Table 2), we found significant differences in intensity (p = 0.004), interference (total, 3.3. Description of longitudinal pain affective, and activity) (p = 0.021; p = 0.033; p = 0.047), and progression medication intake (p = 0.024). The number of pain sites and pain duration instead showed no differences (p = 0.800). Regarding Pain characteristics at baseline in our sample are drawn in the worsening or amelioration of pain intensity and affective Figure 4. The most common location was back pain (44.4%), interference, that is, what we considered as pain experience, 53.3% followed by lower (24.8%) and upper limbs (12.8%). During the showed improvements during this period, and 47.7 % exhibited in-person assessment, participants with CP were asked if they had worsening symptoms. a diagnosis: 29 suffered from migraine, 15 from cervicalgia, 8 from fibromyalgia, 1 from polymyositis traumatic, 16 from knee pathologies, 54 from low back pain, 15 from other diagnoses that 3.4. Chronic pain and the role of brain were cursing with CP and 36 had no specific diagnosis. Several networks in pain progression participants had more than one diagnosis. Regarding the number of pain sites, in the first assessment, When we adjusted the first regression model, where all 75.2% of participants with CP had more than one single painful sociodemographic (BMI, age, and biological sex), health-related site and more than half of the sample had their pain for more than questionnaires (MH, QoL, and cognitive concerns), and pain 3 years (58.7%). The total interference summed mean was 2.66 characteristics (PC, pain medication, intensity, interference, (SD = 2.21), 2.76 (SD = 2.36) in the activity subdomain, and 2.58 duration, and the number of pain sites) were independent variables, (SD = 2.39) in the affective subdomain. Finally, the mean intensity we found that PC (β = 0.481; p < 0.001), pain intensity (β = −0.299; for the last week was 4.17 (SD = 1.47). p = 0.002) and total interference (β = −0.562; p < 0.001) were The pain-relief drugs were the most used treatment at baseline associated with pain experience over time. (31.6%), although 31.6% of participants were not enrolled in The second regression model crucially showed that more segregation of DMN (β = 0.193; p = 0.037), and less segregation TABLE 1 Sociodemographic data and system network segregation in the DAN (β = −0.215; p = 0.014) were associated with differences between no pain and pain group. worst pain experience progression. Moreover, activity and affective interference (β = 0.255; p = 0.022; β = −0.678; p < 0.001, No pain Chronic F p-value (214) pain (133) respectively), pain intensity at baseline (β = −0.428; p < 0.001), and helplessness (β = 0.325; p < 0.003) were also related with worst pain Sociodemographic Mean (SD) Mean (SD) experience progression. Age 52.87 (7.02) 54.81 (7.04) 6,271 0.013* Finally, when we explored the potential moderator role of Biological sex (% 47.2% 63.2 % x2 = 8.396 0.003* helplessness on the relation between SyS and pain experience, we women) found that helplessness moderates the effect of DMN segregation BMI 25.36 (3.72) 26.86 (5.40) 9,362 0.002* on pain experience progression (p = 0.0313, 5,000 bootstrap samples, 95% CI 0.57–1.20), and a tendency in moderate the Health-related questionnaires relation between DAN SyS (p = 0.098, 5,000 bootstrap samples, 95% MH 1.21 (1.59) 2.18 (2.02) 24,772 < 0.001** CI −0.13–1.52) (see Figure 5) and pain experience progression. Cognitive 53.26 (6.54) 48.41 (9.28) 31,945 < 0.001** Johnson-Neyman plots revealed that a score of 7 in the helplessness complaints subscale is the critical value to have a significant association QoL 39.90 (7.93) 34.32 40,392 < 0.001** between DMN segregation and pain experience progression (see Brain networks’ system segregation Figure 6). When we swap the dependent and independent variables Default mode 0.268 (0.098) 2.66 (0.095) 0.831 to explore catastrophism’s helplessness, in interaction with Somatomotor 0.415 (0.082) 0.410 (0.076) 0.871 pain experience, and modulated brain SyS, we found no Control 0.247 (0.084) 0.253 (0.087) 0.380 significant associations. Dorsoattentional 0.305 (0.067) 0.305 (0.066) 0.533 Salience ventral 0.293 (0.077) 0.304 (0.077) 0.117 attentional 4. Discussion Limbic 0.275 (0.104) 0.279 (0.096) 0.822 In this paper, we explored the role of SyS and psychological Visual 0.396 (0.093) 0.393 (0.096) 0.999 aspects in the evolution of pain experience over time. First, Head movement 0.163 (−0.035) 0.179 (0.003) 4,652 0.032* significant differences in self-perceived health status were found *< 0.05; **< 0.001. BMI, body mass index; MH, mental health; QoL, quality of life. between the pain and no pain groups. Second, PC and Frontiers in Neuroscience 08 frontiersin.org Delgado-Gallén et al. 10.3389/fnins.2023.1148176 FIGURE 4 Pain characteristics. (A) Percentages of participants’ pain location. (B) Mean of interference (sum, activity, and affective) and intensity ratings on a numerical rating scale (0–10). (C) Percentages of the number of pain sites. (D) Pain duration of the CP in our sample. brain network segregation of DMN and DAN were the main 4.2. Pain characteristics and their predictors of pain progression. Finally, moderation analyses correlations with psychological aspects revealed that helplessness, a PC’s subdomain, moderated the relationship between DMN and the progression of the pain We found several correlations between pain factors and experience. To our knowledge, this is the first study using the psychological variables (QoL, MH, and self-perceived cognitive fMRI SyS model to identify biomarkers that can predict the complaints), as has been widely described in pain literature progression of CP. (Edwards et al., 2016; Kawai et al., 2017). The subjective and prolonged experience of pain in our cohort was associated with significant mood consequences, such as light depressive 4.1. Differences between groups and symptoms and less capacity to enjoy life. Unquestionably, feeling pain characteristics pain may evolve from normal reactive emotional symptoms, related to stress (Hammen, 2005; Abdallah and Geha, 2017), to When we compared participants with and without CP, clinically relevant depression associated with CP (DeVeaugh-Geiss we found a high prevalence of CP (38.3%) in our sample, et al., 2010). Increasing evidence suggests that the pain-related according to previous reports of a peak prevalence in late plasticity and depression-related neural circuits are responsible middle-aged adults (50–65 years), affecting up to 20–80% of for subtle changes in areas involved in the emotional and people (Gibson and Lussier, 2012). The pain group was older, cognitive aspects of pain over time that contributes to the formed predominantly by women, and had an average BMI behavioral manifestation of altered affective processes (Doan et al., in the overweight range (bordering on mild obesity), also 2015). Finally, as will be discussed later, PC is an important showing the worst health status (MH, cognition, QoL). This is risk factor for a wide range of pain-related effects, including in line with previous studies that found an association between increased pain intensity, increased emotional distress, depression, overweight/obesity and low back pain, particularly stronger in decreased physical function, and prolonged disability (Schütze women than men (Breivik et al., 2006; Stone and Broderick, et al., 2018). 2012; Dahlhamer, 2018). Another difference observed was an increased head motion during the scan for the pain group, possibly related to an increased difficulty to keep still due to their 4.3. Catastrophizing is associated with condition. pain progression We did not observe differences in SyS in any network, probably due to the non-clinical characteristics of our pain sample (i.e., low Our results indicated that changes in pain experience over time disability or low impact in their daily lives, low to mild fluctuations were associated with higher levels of PC and, helplessness was the in pain intensity, and well-controlled pain). only domain that vaticinated greater pain impact. Frontiers in Neuroscience 09 frontiersin.org Delgado-Gallén et al. 10.3389/fnins.2023.1148176 Pain catastrophizing is defined as “an exaggerated negative inhibit thoughts, generate alternative ways of thinking, and switch mental set brought to bear during actual or anticipated painful the focus of attention. PC, which possibly appears as a kind of experience” (Sullivan et al., 2001), and can be divided into three fear of pain, helps perpetuate the cognitive-behavioral pain cycle, main dimensions: magnification (an exaggerated threat value collaborating in the activation of negative cognitive and meta- of pain), rumination (excessive focus pain-related stimuli) and cognitive processes, that in turn can lead to worse coping behaviors helplessness. and an exacerbation of pain (Ziadni et al., 2018). Sullivan (Sullivan, 2012) described how magnification and This is consistent with previous studies focused on cognitive- rumination could overlap with features from primary appraisals affective processes of pain (Gentili et al., 2019; Gonzalez et al., 2019; (i.e., evaluating the pain stimulus), while helplessness overlaps with You et al., 2021). You et al. (2021) demonstrated the existence of secondary appraisal (i.e., the evaluation of oneself to effectively pain-specific resilience, referring to the ability to maintain relatively deal with a stressful situation, like pain stimulus). Consequently, stable and healthy levels of psychological functioning in face of cognitions of helplessness are related to higher perceived pain ongoing and persistent pain (You et al., 2021). Sociodemographic intensity, possibly due to feelings of loss of control, negative future (Tanner et al., 2021), structural and functional MRI (Tanner et al., expectations, or rumination, factors that enhance affective pain 2021; You et al., 2021), clinical pain symptoms, negative pain- experience (Müller, 2011). related emotions, and PC (Sturgeon and Zautra, 2013; Gonzalez Pain catastrophizing is related not only to psychological factors et al., 2019) have been related to resilience and better outcomes in but also to altered brain connectivity in pain-related areas and top- the presence of recurrent pain, while to our knowledge, this is the down inhibition mechanisms [e.g., a reduced engagement of the first study to explore the interaction between all these variables. descending pain modulatory system; see Malfliet et al. (2017) for review]. Rumination in PC can be considered as a cognitive style (“mind-wandering like”) (Èeko et al., 2015), and is responsible for 4.4. System segregation and pain the occurrence of thoughts not related to a given task and not tied to the immediate environment (Häggman-Henrikson et al., 2021). experience progression In general terms, a rumination is a form of circular thinking that We found associations between pain interference and DMN, swallows the individual in a path without a way out, and it can and SN, as well as a significant tendency in DAN. One important be broadly defined as a perseverative self-focused thinking process, point, that distinguishes our results from other studies, is that whereby an individual goes over and over the same thoughts in higher segregation of DMN and lower segregation of DAN were his or her mind. This process, which can be activated during or associated with this pain experience evolution over time, which after a painful event, generally interferes with a person’s ability to is coherent and match with previous results. Besides, helplessness was found to moderate the relationship between DMN and pain TABLE 2 Comparisons between pain variables in first and second questionnaires. experience progression. When we explore the role of the interaction between helplessness and pain experience on brain SyS we found no Pain T1 Pain T2 t p-value significant effects, suggesting that brain SyS in certain networks, in (mean) (mean) interaction with helplessness, affects pain experience evolution over Intensity (last week) 4.17 (1.89) 3.70 (1.95) 2.923 0.004** time, and not the reverse. In this section, we aim to address all the points by substantially Interference total 2.66 (2.21) 2.32 (2.16) 2.332 0.021* revising the main findings on each network individually and Interference affective 2.59 (2.37) 2.23 (2.22) 2.156 0.033* collectively, as well as their relationship with helplessness. Interference activity 2.76 (2.36) 2.44 (2.36) 2.004 0.047* First, concerning DAN, its alteration has already been shown Number of pain sites 2.35 (1.07) 2.38 (1.063) −0.254 0.800 to predict pain intensity progression, possibly due to a cognitive evaluation of pain as a permanent threat to the body (Pfannmöller Duration (< 3 years) 58.7% 36.8% 0.590 0.556 (%) and Lotze, 2019). In this view, our results could be interpreted as dysregulated DAN internal and external connectivity could Duration (3–9 years) 28.6% 36.8% (%) reduce the ability to correctly process nociceptive stimuli and, consequently, increase the perception of pain intensity over time. Duration (> 9 years) 12.7% 26.4% Thus, it could be possible that an imbalance in FC between (%) networks causes a sort of “attentional resources kidnapping,” Medication intake 31.6% 42.9% X2 = 5.115 0.024* increasing the aversive response to pain. This constant painful (% of yes) input and its maladaptive brain changes (i.e., malfunction of the Catastrophizing 6.34 (9.64) descending pain pathways), reinforced by catastrophic thoughts, Helplessness 6.41 (4.19) ultimately provoke an increase in self-perceived pain intensity over Rumination 6.44 (3.37) time. (maximum scoring Similarly, we found a correlation between pain interference 16) and SN segregation, suggesting that, as DAN, altered response in Magnification 3.60 (2.26) this network could be a contributing factor to the maintenance (maximum scoring and chronicity of pain. SN also exhibits task- and resting-state 12) abnormalities in some CP populations (Otti et al., 2013; Becerra *< 0.05; **< 0.001. et al., 2014; Cauda et al., 2014; Qiu et al., 2021), and may be Frontiers in Neuroscience 10 frontiersin.org Delgado-Gallén et al. 10.3389/fnins.2023.1148176 FIGURE 5 System segregation and helplessness predict chronic pain progression. On top, a chart explaining how pain progression has been calculated (subtracting pain experience T1 from pain experience T2). Positive values indicates a worsening while negative results an amelioration. In the middle, two scatter plots reflecting how each network [default mode network (DMN) at left and dorsoattentional network (DAN) at right] was associated with CP progression depending on low/high helplessness (violet, and orange, respectively). At the bottom of the figure, nodes in the graph represent studied regions of interest (ROIs) as defined by the Schaefer-Yeo atlas of 100 nodes and 7 networks. The nodes and edges illustrate ROIs blue for DMN (left) and yellow for DAN (right). Within network connectivity is represented in black (only representative) and refers to outside network ROIs and the connectivity between them and the DAN or DMN. FIGURE 6 Helplessness moderation. At the left, Johnson-Neyman plot show how helplessness moderates the default mode network (DMN) slope since the score 6, where it accentuates the impact of DMN segregation on CP progression. At the right, representative chart of moderation of pain helplessness in the relationship between DMN segregation and pain progression. dysregulated due to constant pain (Borsook et al., 2013), giving rise stimuli, as the SN plays a key role in the assessment of the inherent to a “salient state,” mostly divided into two processes: bottom-up danger of such stimuli (and how one should respond to them), and saliency and top-down control (Melloni et al., 2012). Nonetheless, plays a central role in the memory of painful events (Kim et al., it is well justified that there is an association between a dysregulated 2019). SN and the interplay between pain symptoms and psychological In the same line DMN alteration in CP patients (Malfliet aspects, as several studies have also seen before (Legrain et al., 2011; et al., 2017) has been associated with the “pain state” (Èeko Coppieters et al., 2016; van Ettinger-Veenstra et al., 2019). Indeed, et al., 2020), which facilitates stimulus-independent thoughts, or the function of the SN is essential in the processing of sensory internally directed, spontaneous or autobiographical thoughts, Frontiers in Neuroscience 11 frontiersin.org Delgado-Gallén et al. 10.3389/fnins.2023.1148176 also resumed in the term “self-generated thought” (Andrews- pain experience, and both can be understood as possible Hanna et al., 2014). Repetitive negative thinking is considered a protectors/risk factors. form of avoidant coping strategy (Flink et al., 2013). Occupying Once CP is established, catastrophizing and segregation are not one’s thoughts with repetitive negative thinking prevents the independent factors acting in the pain progression. Segregation confrontation with the threat (e.g., What I could have done to of DMN and DAN, indeed, may represent some aspects of prevent this situation?). Additionally, repetitive negative thinking brain resilience that could help to maintain or get better from can be negatively reinforced by abstract cognitive activity (e.g., Why pain, allowing the brain to effectively process pain, or at least do I suffer from pain?). This form of abstract thinking impedes avoid areas not involved in pain processing. On the other hand, the activation and processing of emotional and somatic responses. modifiable psychologic aspects like catastrophizing, related to The process of suppression/avoidance could magnify the negative cognitive reserve, may have a crucial role in pain evolution emotions and consequently fuel the catastrophic worry cycle. and the impact of MH. Both kinds of processes can be treated Usually, some of these thoughts are related to personal significance, separately but in a coordinated way, as seen in studies of combined temporal or social orientation, or somatosensory awareness. Ergo, behavioral therapies (e.g., pain neuroscience education) with DMN seems to play an important active role in the self-generation brain stimulation (Meeker et al., 2020; Alcon and Wang-Price, of cognition, thinking about one-self and future thinking, but also 2022), and could reduce the negative impact of emotions in the with rumination and PC (Èeko et al., 2020). evolution of pain over time, collaborating in the selection of The predictive value of DMN rs-FC on pain management effective coping strategies or avoiding negative thinking. Thereby, was already shown by Baliki et al. (2014), where CP disrupted descending modulation might be more effective, decreasing levels the dynamics of DMN and frontoparietal network (related to of pain intensity and affective interference. Finally, resilience- attention and working memory), and other regions related to focused cognitive-affective approaches, including this holistic view, pain modulation (insula; anterior cingulate cortex, ACC) (Baliki are important to be applied not only in highly affected CP patients, et al., 2014). Specific nodes of DMN, like the hippocampus and but also in individuals with a low rating in pain parameters, as they medial prefrontal cortex (mPFC), are importantly implicated in can already show lower QoL, cognition, and MH, or FC alterations. pain processing. Similarly, Kucyi et al. (2014) demonstrated an enhanced mPFC connectivity in DMN, suggesting its role in the descending modulatory system underlying the degree to which patients ruminate about their CP (Kucyi et al., 2014). Hashmi et al. 4.6. Limitations and future directions (2013) also found that strengthening mPFC-nucleus accumbens FC predicts the extension in which brain activity shifts from Our main limitation was that we investigated a non- pain-related to emotion-related regions in patients with persistent homogeneous (e.g., different kinds of pain) and non-clinical sample subacute back pain, compared to those that get total recovery (low ratings in pain intensity, interference, and PC). Thus, in (Hashmi et al., 2013). In sum, the mPFC can be seen as a hub future work, investigating how higher catastrophizing scores, as for the development of mental comorbidities associated with CP. well as intensity or interference ratings, interact with networks’ The impaired cholinergic activity contributes to the deactivation of segregation in the progression of pain might shed light on how mPFC, possibly leading to cognitive and emotional deficits in CP pain alters different networks and how it’s linked to emotional and patients (Kummer et al., 2020). cognitive aspects. Overall, our results are consistent with findings reported by Ter Another limitation of our study was the lack of information Minassian et al. (2012) in their study about brain activity during about pain during MRI or the presence of acute pain in the no- pain anticipation and pain perception (Ter Minassian et al., 2012). pain sample. Further work is certainly required to disentangle They found that during pain anticipation, DAN was activated while these complexities in acute (non-experimentally provoked) and DMN was deactivated, possibly due to the search for strategies CP. For example, the possibility of prognostic biomarkers of the to avoid pain. In contrast, during pain perception, DMN was reactivated, whereas DAN remained activated, supporting the fact transition from acute pain to CP, within the framework of pain- that DMN and attentional networks cooperate to integrate pain- related resilience (SyS and psychological factors) is described in related stimuli and thoughts, as DMN is at the top of the networks this manuscript. that perform hierarchical integration. The authors concluded that An additional potential limitation was related to time gaps the ACC could be the structure most involved in the coordination between questionnaires and MRI. Even though we used this gap as between these two networks, given that it can be anatomically and a covariable in our analysis, we cannot exclude that this aspect can functionally divided during pain anticipation (cognitive; caudal introduce some bias in the results. ACC) and pain perception (emotional; rostral ACC). Although we excluded subjects with excess motion and added the mean head motion of each participant as a covariate in all the analyses related to fMRI, we cannot fully dismiss that subject motion to some degree has influenced the present results. 4.5. Default mode and dorsoattentional Finally, due to the exploratory nature of our analyses, networks interactions with helplessness we did not correct functional connectivity analysis for multiple comparisons. Our results indicate that brain network functioning and Future studies need to confirm these results with better cognitive-emotional strategies can interact in modulating controlling for all these aspects and overcome present limitations. Frontiers in Neuroscience 12 frontiersin.org Delgado-Gallén et al. 10.3389/fnins.2023.1148176 5. Conclusion made substantial contributions to analysis and interpretation of data. GE-I, AR-V, MC-T, KA-P, JS-S, and SD-G have made In summary, present results indicate that non-clinical CP substantial contribution to acquisition of data. MDS, GC, and SD-G conditions are associated with the segregation of brain networks participated in drafting the manuscript. AP-L, DB-F, JMT, GE-I, during rs-fMRI, more concretely SN, DAN, and DMN. Moreover, AR-V, MC-T, KA-P, and JS-S contributed to revising it critically for the latter effect is moderated by helplessness, a catastrophizing important intellectual content. All authors have given final approval domain, that can be seen as a protective factor against pain of the version to be submitted. impact over time. Our result casts a new light on the interplay among Funding the influence of the interaction between psychological aspects and brain networks in pain management. The extent of the This research has received funding from La Caixa Foundation reorganization of these networks and their functioning is critical (grant number LCF/PR/PR16/11110004), Institut Guttmann, and for the evolution of pain. Fundació Abertis. AP-L was partially supported by Fundació La Besides, the reorganization of spatial properties of the DMN, Marató De TV3 (202129-30-31). DB-F was supported in part DAN, and SN may reflect different emotional, attentional, and by the Spanish Ministry of Science, Innovation and Universities cognitive abnormalities observed in CP conditions. Under certain (MICIU/FEDER; grant number RTI2018-095181-B-C21) and the assumptions, this can be construed as the common reorganization Catalan Institute for Advanced Research and Studies Academia among pain patients is the extent of association of the component of 2019 grant award and Fundació La Marató De TV3 (202129- the DAN, and its dissociation from the posterior components of the 30-31). JMT was partly supported by the Fundació Joan DMN, which seems to disrupt the competitive inhibition between Ribas Araquistain_Fjra (PREHABILITA-IUN91752), Agency for the DMN and the brain networks underlying attention. Management of University and Research Grants, Agència de Gestió Finally, our results highlight the predictive value of fMRI d’Ajuts Universitaris i de Recerca (2018 PROD 00172), and the that can orient researchers and clinicians about what form of European Commission (call H2020-SC1-2016-2017_RIA_777107). treatable pathophysiology an individual patient with CP may GC was partly supported by the Fundació La Marató De TV3 have. Given that other fMRI techniques are not suitable to (202211-30-31). KA-P was financially supported by a Juan de measure certain characteristics, such as the effect of attentional la Cierva-Formación research grant (FJC2021-047380-I) of the fluctuation or the degree of self-generated thoughts, we conclude Spanish Ministry of Science and Innovation. that SyS networks are an effective metric to understand brain connectivity and neural correlates of the pain state. SyS could be used as biomarkers indicating a predisposition of pain maintenance Acknowledgments and allowing disease progression monitoring, as well as the reversion or compensation of these alterations in different kinds of Special thanks are extended to all participants, the project interventions, that should, in the future, be considered for more team and research assistants, and other partners (Ad-Salutem deep research. We have demonstrated that these variables should Institute, Sodexo, I.C.A. Informàtica y Comunicaciones Avanzadas, be assessed without considering the etiology of pain, giving more Neuroelectrics, Corporació Catalana de Mitjans Audiovisuals, Club importance to the pain state than the pain condition per se. Metropolitan, Casa Ametller, and Agència de Qualitat i Avaluació Sanitàries de Catalunya-AQuAS) for their invaluable collaboration. Data availability statement Conflict of interest The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation. The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. Ethics statement AP-L received funding from MagStim Inc. The funder was not involved in the study design, collection, analysis, interpretation of The protocol was approved by the “Comité d’Ètica i data, the writing of this article, or the decision to submit It for Investigació Clínica de la Unió Catalana d’Hospitals” and was publication. carried out following the Declaration of Helsinki (World Medical Association, 2013). Written informed consent was obtained from all participants before inclusion in the study. Publisher’s note All claims expressed in this article are solely those of the Author contributions authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the MDS, GC, GE-I, AR-V, MC-T, KA-P, JS-S, AP-L, DB-F, JMT, reviewers. Any product that may be evaluated in this article, or and SD-G have made substantial contributions to conception, claim that may be made by its manufacturer, is not guaranteed or design, and interpretation of data. GC, MDS, and SD-G have endorsed by the publisher. Frontiers in Neuroscience 13 frontiersin.org Delgado-Gallén et al. 10.3389/fnins.2023.1148176 References Abdallah, C. G., and Geha, P. (2017). Chronic pain and chronic stress: two sides health in middle-aged adults affected by chronic pain. Front. Psychol. 12:4557. doi: of the same coin? Chronic Stress 1:2470547017704763. doi: 10.1177/247054701770 10.3389/fpsyg.2021.752623 4763 DeVeaugh-Geiss, A. M., West, S. L., Miller, W. C., Sleath, B., Gaynes, B. N., and Abeler, K., Friborg, O., Engstrøm, M., Sand, T., and Bergvik, S. (2020). Sleep Kroenke, K. (2010). The adverse effects of comorbid pain on depression outcomes characteristics in adults with and without chronic musculoskeletal pain: the role of in primary care patients: results from the ARTIST trial. Pain Med. 11, 732–741. mental distress and pain catastrophizing. Clin. J. Pain 36, 707–715. doi: 10.1097/AJP. doi: 10.1111/j.1526-4637.2010.00830.x 0000000000000854 Doan, L., Manders, T., and Wang, J. (2015). Neuroplasticity underlying the Alcon, C. A., and Wang-Price, S. (2022). Non-invasive brain stimulation and pain comorbidity of pain and depression. Neural Plast. 2015:504691. doi: 10.1155/2015/ neuroscience education in the cognitive-affective treatment of chronic low back pain: 504691 evidence and future directions. Front. Pain Res. 3:959609. doi: 10.3389/fpain.2022. Dong, H.-J., Gerdle, B., Bernfort, L., Levin, L. -Å, and Dragioti, E. (2020). Pain 959609 catastrophizing in older adults with chronic pain: the mediator effect of mood using a Andrews-Hanna, J. R., Smallwood, J., and Spreng, R. N. (2014). The default network path analysis approach. J. Clin. Med. 9:2073 doi: 10.3390/jcm9072073 and self-generated thought: component processes, dynamic control, and clinical Edwards, R. R., Dworkin, R. H., Sullivan, M. D., Turk, D. C., and Wasan, A. D. relevance. Ann. N. Y. Acad. Sci. 1316, 29–52. doi: 10.1111/nyas.12360 (2016). The role of psychosocial processes in the development and maintenance of Badia, X., Muriel, C., Gracia, A., Manuel Núñez-Olarte, J., Perulero, N., Gálvez, R., chronic pain. J. Pain 17, T70–T92. doi: 10.1016/j.jpain.2016.01.001 et al. (2003). Validación española del cuestionario brief pain inventory en pacientes con Ewers, M., Luan, Y., Frontzkowski, L., Neitzel, J., Rubinski, A., Dichgans, M., dolor de causa neoplásica. Med. Clín. 120, 52–59. doi: 10.1016/S0025-7753(03)73601- et al. (2021). Segregation of functional networks is associated with cognitive X resilience in Alzheimer’s disease. Brain 144, 2176–2185. doi: 10.1093/brain/awa Baliki, M. N., Mansour, A. R., Baria, A. T., and Apkarian, A. V. (2014). Functional b112 reorganization of the default mode network across chronic pain conditions. PLoS One Fieo, R., Ocepek-Welikson, K., Kleinman, M., Eimicke, J. P., Crane, P. K., Cella, D., 9:e106133. doi: 10.1371/journal.pone.0106133 et al. (2016). Measurement equivalence of the patient reported outcomes measurement Becerra, L., Sava, S., Simons, L. E., Drosos, A. M., Sethna, N., Berde, C., et al. (2014). information system R (PROMIS R ) applied cognition – general concerns, short forms Intrinsic brain networks normalize with treatment in pediatric complex regional pain in ethnically diverse groups. Psychol. Test Assess Model. 58, 255–307. syndrome. Neuroimage Clin. 6, 347–369. doi: 10.1016/j.nicl.2014.07.012 Fischl, B., Salat, D. H., Busa, E., Albert, M., Dieterich, M., Haselgrove, C., et al. Borsook, D., Edwards, R., Elman, I., Becerra, L., and Levine, J. (2013). Pain and (2002). Whole brain segmentation: automated labeling of neuroanatomical structures analgesia: the value of salience circuits. Prog. Neurobiol. 104, 93–105. doi: 10.1016/j. in the human brain. Neuron 33, 341–355. doi: 10.1016/S0896-6273(02)00569-X pneurobio.2013.02.003 Flink, I. L., Boersma, K., and Linton, S. J. (2013). Pain catastrophizing as repetitive Breivik, H., Collett, B., Ventafridda, V., Cohen, R., and Gallacher, D. (2006). Survey negative thinking: a development of the conceptualization. Cogn. Behav. Ther. 42, of chronic pain in Europe: prevalence, impact on daily life, and treatment. Eur. J. Pain 215–223. doi: 10.1080/16506073.2013.769621 10, 287–333. doi: 10.1016/j.ejpain.2005.06.009 Franzmeier, N., Göttler, J., Grimmer, T., Drzezga, A., Áraque-Caballero, M. A., Burns, L. C., Ritvo, S. E., Ferguson, M. K., Clarke, H., Seltzer, Z., and Katz, J. (2015). Simon-Vermot, L., et al. (2017). Resting-state connectivity of the left frontal cortex Pain catastrophizing as a risk factor for chronic pain after total knee arthroplasty: a to the default mode and dorsal attention network supports reserve in mild cognitive systematic review. J. Pain Res. 8, 21–32. doi: 10.2147/JPR.S64730 impairment. Front. Aging Neurosci. 9:264. doi: 10.3389/fnagi.2017.00264 Cattaneo, G., Bartrés-Faz, D., Morris, T. P., Sánchez, J. S., Macià, D., Tarrero, García Campayo, J., Rodero, B., Alda, M., Sobradiel, N., Montero, J., and Moreno, C., et al. (2018). The barcelona brain health initiative: a cohort study to define and S. (2008). [Validation of the Spanish version of the pain catastrophizing scale in promote determinants of brain health. Front. Aging Neurosci. 10:321. doi: 10.3389/ fibromyalgia]. Med. Clin. 131, 487–492. doi: 10.1157/13127277 fnagi.2018.00321 Gentili, C., Rickardsson, J., Zetterqvist, V., Simons, L. E., Lekander, M., and Wicksell, Cattaneo, G., Bartrés-Faz, D., Morris, T. P., Sánchez, J. S., Macià, D., Tormos, R. K. (2019). Psychological flexibility as a resilience factor in individuals with chronic J. M., et al. (2020). The barcelona brain health initiative: cohort description and first pain. Front. Psychol. 10:2016. doi: 10.3389/fpsyg.2019.02016 follow-up. PLoS one 15:e0228754. doi: 10.1371/journal.pone.0228754 Gibson, S. J., and Lussier, D. (2012). Prevalence and relevance of pain in older Cauda, F., Palermo, S., Costa, T., Torta, R., Duca, S., Vercelli, U., et al. (2014). persons. Pain Med. 13, S23–S26. doi: 10.1111/j.1526-4637.2012.01349.x Gray matter alterations in chronic pain: a network-oriented meta-analytic approach. Gilliam, W. P., Craner, J. R., Morrison, E. J., and Sperry, J. A. (2017). The Neuroimage Clin. 4, 676–686. doi: 10.1016/j.nicl.2014.04.007 mediating effects of the different dimensions of pain catastrophizing on outcomes Èeko, M., Frangos, E., Gracely, J., Richards, E., Wang, B., Schweinhardt, P., et al. in an interdisciplinary pain rehabilitation program. Clin. J. Pain 33, 443–451. doi: (2020). Default mode network changes in fibromyalgia patients are largely dependent 10.1097/AJP.0000000000000419 on current clinical pain. NeuroImage 216:116877. doi: 10.1016/j.neuroimage.2020. Gonzalez, C. E., Okunbor, J. I., Parker, R., Owens, M. A., White, D. M., Merlin, 116877 J. S., et al. (2019). Pain-specific resilience in people living with HIV and chronic Èeko, M., Gracely, J. L., Fitzcharles, M.-A., Seminowicz, D. A., Schweinhardt, pain: beneficial associations with coping strategies and catastrophizing. Front. Psychol. P., and Bushnell, M. C. (2015). Is a responsive default mode network required for 10:2046. doi: 10.3389/fpsyg.2019.02046 successful working memory task performance? J. Neurosci. 35, 11595–11605. doi: Haefeli, M., and Elfering, A. (2006). Pain assessment. Eur. Spine J. 15, S17–S24. 10.1523/JNEUROSCI.0264-15.2015 doi: 10.1007/s00586-005-1044-x Cleeland, C. S. (2009). The M. D. Anderson symptom inventory. User guide-version Häggman-Henrikson, B., Visscher, C. M., Wänman, A., Ljótsson, B., Peck, C. C., 1. Available online at: https://www.mdanderson.org/documents/Departments-and- and Lövgren, A. (2021). Even mild catastrophic thinking is related to pain intensity in Divisions/Symptom-Research/MDASI_userguide.pdf individuals with painful temporomandibular disorders. J. Oral Rehabil. 48, 1193–1200. Coppieters, I., Meeus, M., Kregel, J., Caeyenberghs, K., De Pauw, R., Goubert, D., doi: 10.1111/joor.13251 et al. (2016). Relations between brain alterations and clinical pain measures in chronic Hammen, C. (2005). Stress and depression. Annu. Rev. Clin. Psychol. 1, 293–319. musculoskeletal pain: a systematic review. J. Pain 17, 949–962. doi: 10.1016/j.jpain. doi: 10.1146/annurev.clinpsy.1.102803.143938 2016.04.005 Hashmi, J. A., Baliki, M. N., Huang, L., Baria, A. T., Torbey, S., Hermann, K. M., Coppola, G., Di Renzo, A., Petolicchio, B., Tinelli, E., Di Lorenzo, C., Parisi, V., et al. et al. (2013). Shape shifting pain: chronification of back pain shifts brain representation (2019). Aberrant interactions of cortical networks in chronic migraine: a resting-state from nociceptive to emotional circuits. Brain 136, 2751–2768. doi: 10.1093/brain/ fMRI study. Neurology 92, e2550–e2558. doi: 10.1212/WNL.0000000000007577 awt211 Craner, J. R., Gilliam, W. P., and Sperry, J. A. (2016). Rumination, magnification, Hemington, K. S., Wu, Q., Kucyi, A., Inman, R. D., and Davis, K. D. (2016). and helplessness: how do different aspects of pain catastrophizing relate to Abnormal cross-network functional connectivity in chronic pain and its association pain severity and functioning? Clin. J. Pain 32, 1028–1035. doi: 10.1097/AJP. with clinical symptoms. Brain Struct. Funct. 221, 4203–4219. doi: 10.1007/s00429-015- 0000000000000355 1161-1 Dahlhamer, J. (2018). Prevalence of chronic pain and high-impact chronic pain Jensen, M. P., Tomé-Pires, C., de la Vega, R., Galán, S., Solé, E., and Miró, J. (2017). among adults — United States, 2016. MMWR Morb Mortal Wkly Rep. 67, 1001–1006. What determines whether a pain is rated as mild, moderate, or severe? the importance doi: 10.15585/mmwr.mm6736a2 of pain beliefs and pain interference. Clin. J. Pain 33, 414–421. doi: 10.1097/AJP. De Ridder, D., Vanneste, S., Smith, M., and Adhia, D. (2022). Pain and the triple 0000000000000429 network model. Front. Neurol. 13:757241. doi: 10.3389/fneur.2022.757241 Jones, S. A., Morales, A. M., Holley, A. L., Wilson, A. C., and Nagel, B. J. (2020). Delgado-Gallén, S., Soler, M. D., Albu, S., Pachón-García, C., Alviárez-Schulze, V., Default mode network connectivity is related to pain frequency and intensity in Solana-Sánchez, J., et al. (2021). Cognitive reserve as a protective factor of mental adolescents. Neuroimage Clin. 27:102326. doi: 10.1016/j.nicl.2020.102326 Frontiers in Neuroscience 14 frontiersin.org Delgado-Gallén et al. 10.3389/fnins.2023.1148176 Kastrati, G., Thompson, W. H., Schiffler, B., Fransson, P., and Jensen, K. B. (2022). Otti, A., Guendel, H., Wohlschläger, A., Zimmer, C., and Noll-Hussong, M. (2013). Brain network segregation and integration during painful thermal stimulation. Cereb. Frequency shifts in the anterior default mode network and the salience network in Cortex 32, 4039–4049. doi: 10.1093/cercor/bhab464 chronic pain disorder. BMC Psychiatry 13:84. doi: 10.1186/1471-244X-13-84 Kawai, K., Kawai, A. T., Wollan, P., and Yawn, B. P. (2017). Adverse impacts Pfannmöller, J., and Lotze, M. (2019). Review on biomarkers in the resting-state of chronic pain on health-related quality of life, work productivity, depression and networks of chronic pain patients. Brain Cogn. 131, 4–9. doi: 10.1016/j.bandc.2018.06. anxiety in a community-based study. Fam. Pract. 34, 656–661. doi: 10.1093/fampra/ 005 cmx034 Power, J. D., Barnes, K. A., Snyder, A. Z., Schlaggar, B. L., and Petersen, S. E. (2012). Keller, S., Bann, C. M., Dodd, S. L., Schein, J., Mendoza, T. R., and Cleeland, C. S. Spurious but systematic correlations in functional connectivity MRI networks arise (2004). Validity of the brief pain inventory for use in documenting the outcomes from subject motion. Neuroimage 59, 2142–2154. doi: 10.1016/j.neuroimage.2011.10. of patients with noncancer pain. Clin. J. Pain 20, 309–318. doi: 10.1097/00002508- 018 200409000-00005 Power, J. D., Barnes, K. A., Snyder, A. Z., Schlaggar, B. L., and Petersen, S. E. (2013). Kilpatrick, L. A., Istrin, J. J., Gupta, A., Naliboff, B., Tillisch, K., Labus, J. S., Steps toward optimizing motion artifact removal in functional connectivity MRI; a et al. (2015). Sex commonalities and differences in the relationship between resilient reply to carp. Neuroimage 76, 439–441. doi: 10.1016/j.neuroimage.2012.03.017 personality and the intrinsic connectivity of the salience and default mode networks. Power, J. D., Mitra, A., Laumann, T. O., Snyder, A. Z., Schlaggar, B. L., and Petersen, Biol. Psychol. 112, 107–115. doi: 10.1016/j.biopsycho.2015.09.010 S. E. (2014). Methods to detect, characterize, and remove motion artifact in resting Kim, J., Mawla, I., Kong, J., Lee, J., Gerber, J., Ortiz, A., et al. (2019). Somatotopically state fMRI. Neuroimage 84, 320–341. doi: 10.1016/j.neuroimage.2013.08.048 specific primary somatosensory connectivity to salience and default mode networks Power, J. D., Schlaggar, B. L., and Petersen, S. E. (2015). Recent progress and encodes clinical pain. Pain 160, 1594–1605. doi: 10.1097/j.pain.0000000000001541 outstanding issues in motion correction in resting state fMRI. Neuroimage 105, Kocalevent, R.-D., Finck, C., Jimenez-Leal, W., Sautier, L., and Hinz, A. (2014). 536–551. doi: 10.1016/j.neuroimage.2014.10.044 Standardization of the Colombian version of the PHQ-4 in the general population. Qiu, J., Du, M., Yang, J., Lin, Z., Qin, N., Sun, X., et al. (2021). The brain’s structural BMC Psychiatry 14:205. doi: 10.1186/1471-244X-14-205 differences between postherpetic neuralgia and lower back pain. Sci. Rep. 11:22455. Kucyi, A., Moayedi, M., Weissman-Fogel, I., Goldberg, M. B., Freeman, B. V., doi: 10.1038/s41598-021-01915-x Tenenbaum, H. C., et al. (2014). Enhanced medial prefrontal-default mode network Riedel, L., van den Heuvel, M. P., and Markett, S. (2021). Trajectory of rich club functional connectivity in chronic pain and its association with pain rumination. properties in structural brain networks. Hum. Brain Mapp. 43, 4239–4253. doi: 10. J. Neurosci. 34, 3969–3975. doi: 10.1523/JNEUROSCI.5055-13.2014 1002/hbm.25950 Kummer, K. K., Mitriæ, M., Kalpachidou, T., and Kress, M. (2020). The medial Schaefer, A., Kong, R., Gordon, E. M., Laumann, T. O., Zuo, X.-N., Holmes, A. J., prefrontal cortex as a central hub for mental comorbidities associated with chronic et al. (2018). Local-global parcellation of the human cerebral cortex from intrinsic pain. Int. J. Mol. Sci. 21, E3440. doi: 10.3390/ijms21103440 functional connectivity MRI. Cereb. Cortex 28, 3095–3114. doi: 10.1093/cercor/bh Lee, J.-J., Lee, S., Lee, D. H., and Woo, C.-W. (2022). Functional brain x179 reconfiguration during sustained pain. eLife 11:e74463. doi: 10.7554/eLife.74463 Schütze, R., Rees, C., Smith, A., Slater, H., Campbell, J. M., and O’Sullivan, P. (2018). Legrain, V., Damme, S. V., Eccleston, C., Davis, K. D., Seminowicz, D. A., and How can we best reduce pain catastrophizing in adults with chronic noncancer pain? Crombez, G. (2009). A neurocognitive model of attention to pain: behavioral and a systematic review and meta-analysis. J. Pain 19, 233–256. doi: 10.1016/j.jpain.2017. neuroimaging evidence. Pain 144, 230–232. doi: 10.1016/j.pain.2009.03.020 09.010 Legrain, V., Iannetti, G. D., Plaghki, L., and Mouraux, A. (2011). The pain matrix Seidler, R., Erdeniz, B., Koppelmans, V., Hirsiger, S., Mérillat, S., and Jäncke, L. reloaded: a salience detection system for the body. Prog. Neurobiol. 93, 111–124. (2015). Associations between age, motor function, and resting state sensorimotor doi: 10.1016/j.pneurobio.2010.10.005 network connectivity in healthy older adults. NeuroImage 108, 47–59. doi: 10.1016/ j.neuroimage.2014.12.023 Loggia, M. L., Kim, J., Gollub, R. L., Vangel, M. G., Kirsch, I., Kong, J., et al. (2013). Default mode network connectivity encodes clinical pain: an arterial spin labeling Seminowicz, D. A., Wideman, T. H., Naso, L., Hatami-Khoroushahi, Z., Fallatah, study. Pain 154, 24–33. doi: 10.1016/j.pain.2012.07.029 S., Ware, M. A., et al. (2011). Effective treatment of chronic low back pain in humans reverses abnormal brain anatomy and function. J. Neurosci. 31, 7540–7550. doi: 10. Lucas-Carrasco, R. (2012). The WHO quality of life (WHOQOL) questionnaire: 1523/JNEUROSCI.5280-10.2011 Spanish development and validation studies. Qual. Life Res. 21, 161–165. doi: 10.1007/ s11136-011-9926-3 Solé-Padullés, C., Cattaneo, G., Marchant, N. L., Cabello-Toscano, M., Mulet-Pons, L., Solana, J., et al. (2022). Associations between repetitive negative thinking and Malagurski, B., Liem, F., Oschwald, J., Mérillat, S., and Jäncke, L. (2020). Functional resting-state network segregation among healthy middle-aged adults. Front. Aging dedifferentiation of associative resting state networks in older adults – A longitudinal Neurosci. 14:1062887. doi: 10.3389/fnagi.2022.1062887 study. NeuroImage 214:116680. doi: 10.1016/j.neuroimage.2020.116680 Spisak, T., Kincses, B., Schlitt, F., Zunhammer, M., Schmidt-Wilcke, T., Kincses, Malfliet, A., Coppieters, I., Van Wilgen, P., Kregel, J., De Pauw, R., Dolphens, M., Z. T., et al. (2019). Pain-free resting-state functional brain connectivity predicts et al. (2017). Brain changes associated with cognitive and emotional factors in chronic individual pain sensitivity. bioRxiv [Preprint] doi: 10.1101/790709 pain: a systematic review. Eur. J. Pain 21, 769–786. doi: 10.1002/ejp.1003 Spreng, R. N., Sepulcre, J., Turner, G. R., Stevens, W. D., and Schacter, D. L. (2013). Mao, C. P., Yang, H. J., Zhang, Q. J., Yang, Q. X., and Li, X. H. (2022). Altered Intrinsic architecture underlying the relations among the default, dorsal attention, effective connectivity within the cingulo-frontal-parietal cognitive attention networks and frontoparietal control networks of the human brain. J. Cogn. Neurosci. 25, 74–86. in chronic low back pain: a dynamic causal modeling study. Brain Imaging Behav. 16, doi: 10.1162/jocn_a_00281 1516–1527. doi: 10.1007/s11682-021-00623-4 Stensland, M. (2021). “If you don’t keep going, you’re gonna die”: helplessness and Meeker, T. J., Jupudi, R., Lenz, F. A., and Greenspan, J. D. (2020). New developments perseverance among older adults living with chronic low back pain. Gerontologist 61, in non-invasive brain stimulation in chronic pain. Curr. Phys. Med. Rehabil. Rep. 8, 907–916. doi: 10.1093/geront/gnaa150 280–292. doi: 10.1007/s40141-020-00260-w Stone, A. A., and Broderick, J. E. (2012). Obesity and pain are associated in the Melloni, L., van Leeuwen, S., Alink, A., and Müller, N. G. (2012). Interaction United States. Obesity 20, 1491–1495. doi: 10.1038/oby.2011.397 between bottom-up saliency and top-down control: how saliency maps are created in the human brain. Cereb. Cortex 22, 2943–2952. doi: 10.1093/cercor/bhr384 Sturgeon, J. A., and Zautra, A. J. (2013). Psychological resilience, pain catastrophizing, and positive emotions: perspectives on comprehensive modeling of Miettinen, T., Kautiainen, H., Mäntyselkä, P., Linton, S. J., and Kalso, E. (2019). Pain individual pain adaptation. Curr. Pain Headache Rep. 17:317. doi: 10.1007/s11916- interference type and level guide the assessment process in chronic pain: categorizing 012-0317-4 pain patients entering tertiary pain treatment with the brief pain inventory. PLoS One 14:e0221437. doi: 10.1371/journal.pone.0221437 Sullivan, M. J., Thorn, B., Haythornthwaite, J. A., Keefe, F., Martin, M., Bradley, L. A., et al. (2001). Theoretical perspectives on the relation between catastrophizing Müller, M. J. (2011). Helplessness and perceived pain intensity: relations to cortisol and pain. Clin. J. Pain 17, 52–64. doi: 10.1097/00002508-200103000-00008 concentrations after electrocutaneous stimulation in healthy young men. Biopsychosoc. Med. 5:8. doi: 10.1186/1751-0759-5-8 Sullivan, M. J. L. (2012). The communal coping model of pain catastrophising: clinical and research implications. Can. Psychol. 53, 32–41. doi: 10.1037/a0026726 Napadow, V., LaCount, L., Park, K., As-Sanie, S., Clauw, D. J., and Harris, R. E. (2010). Intrinsic brain connectivity in fibromyalgia is associated with chronic pain Sullivan, M. J. L., Bishop, S. R., and Pivik, J. (1995). The pain catastrophizing scale: intensity. Arthritis Rheum. 62, 2545–2555. doi: 10.1002/art.27497 development and validation. Psychol. Assess. 7, 524–532. doi: 10.1037/1040-3590.7.4. 524 Olmedilla Zafra, A., Ortega Toro, E., and Abenza Cano, L. (2013). Validación de la escala de catastrofismo ante el dolor (pain catastrophizing scale) en deportistas Suso-Ribera, C., García-Palacios, A., Botella, C., and Ribera-Canudas, M. V. españoles. Cuadernos de Psicología del Deporte 13, 83–94. doi: 10.4321/S1578- (2017). Pain catastrophizing and its relationship with health outcomes: does 84232013000100009 pain intensity matter? Pain Res. Manag. 2017:9762864. doi: 10.1155/2017/976 2864 Opdebeeck, C., Matthews, F. E., Wu, Y.-T., Woods, R. T., Brayne, C., and Clare, L. (2018). Cognitive reserve as a moderator of the negative association between mood Tanner, J. J., Johnson, A. J., Terry, E. L., Cardoso, J., Garvan, C., Staud, R., et al. and cognition: evidence from a population-representative cohort. Psychol. Med. 48, (2021). Resilience, pain, and the brain: relationships differ by sociodemographics. 61–71. doi: 10.1017/S003329171700126X J. Neurosci. Res. 99, 1207–1235. doi: 10.1002/jnr.24790 Frontiers in Neuroscience 15 frontiersin.org Delgado-Gallén et al. 10.3389/fnins.2023.1148176 Ter Minassian, A., Ricalens, E., Humbert, S., Duc, F., Aubé, C., and Beydon, L. Wiech, K. (2016). Deconstructing the sensation of pain: the influence of cognitive (2012). Dissociating anticipation from perception: Acute pain activates default mode processes on pain perception. Science 354, 584–587. doi: 10.1126/science.aaf8934 network. Hum. Brain Mapp. 34, 2228–2243. doi: 10.1002/hbm.22062 Wiech, K. (2018). Biased perception and learning in pain. Nat. Hum. Behav. 2, The Whoqol Group (1998). The world health organization quality of life assessment 804–805. doi: 10.1038/s41562-018-0468-3 (WHOQOL): development and general psychometric properties. Soc. Sci. Med. 46, Wiech, K., and Shriver, A. (2018). Cognition doesn’t only modulate pain perception; 1569–1585. doi: 10.1016/S0277-9536(98)00009-4 it’s a central component of it. AJOB Neurosci. 9, 196–198. Thomas Yeo, B. T., Krienen, F. M., Sepulcre, J., Sabuncu, M. R., Lashkari, D., Wig, G. S. (2017). Segregated systems of human brain networks. Trends Cogn. Sci. Hollinshead, M., et al. (2011). The organization of the human cerebral cortex estimated 21, 981–996. doi: 10.1016/j.tics.2017.09.006 by intrinsic functional connectivity. J. Neurophysiol. 106, 1125–1165. doi: 10.1152/jn. 00338.2011 World Medical Association (2013). World medical association declaration of Helsinki: Ethical principles for medical research involving human subjects. JAMA 310, Treede, R.-D., Rief, W., Barke, A., Aziz, Q., Bennett, M. I., Benoliel, R., et al. (2015). 2191–2194. doi: 10.1001/jama.2013.281053 A classification of chronic pain for ICD-11. Pain 156, 1003–1007. doi: 10.1097/j.pain. 0000000000000160 Yoshino, A., Okamoto, Y., Okada, G., Takamura, M., Ichikawa, N., Shibasaki, C., et al. (2018). Changes in resting-state brain networks after cognitive-behavioral van den Heuvel, M. P., and Sporns, O. (2013). An anatomical substrate for therapy for chronic pain. Psychol. Med. 48, 1148–1156. integration among functional networks in human cortex. J. Neurosci. 33, 14489–14500. doi: 10.1523/JNEUROSCI.2128-13.2013 You, B., Wen, H., and Jackson, T. (2021). Identifying resting state differences salient for resilience to chronic pain based on machine learning Van Dijk, K. R. A., Sabuncu, M. R., and Buckner, R. L. (2012). The influence multivariate pattern analysis. Psychophysiology 58:e13921. doi: 10.1111/psyp.1 of head motion on intrinsic functional connectivity MRI. Neuroimage 59, 431–438. 3921 doi: 10.1016/j.neuroimage.2011.07.044 Yu, R., Gollub, R. L., Spaeth, R., Napadow, V., Wasan, A., and Kong, J. (2014). van Ettinger-Veenstra, H., Lundberg, P., Alföldi, P., Södermark, M., Graven- Disrupted functional connectivity of the periaqueductal gray in chronic low back pain. Nielsen, T., Sjörs, A., et al. (2019). Chronic widespread pain patients show Neuroimage Clin. 6, 100–108. doi: 10.1016/j.nicl.2014.08.019 disrupted cortical connectivity in default mode and salience networks, Zhaoyang, R., Martire, L. M., and Darnall, B. D. (2020). Daily pain catastrophizing modulated by pain sensitivity. J. Pain Res. 12, 1743–1755. doi: 10.2147/JPR.S18 predicts less physical activity and more sedentary behavior in older adults 9443 with osteoarthritis. Pain 161, 2603–2610. doi: 10.1097/j.pain.000000000000 Vossel, S., Geng, J. J., and Fink, G. R. (2014). Dorsal and ventral attention systems. 1959 Neuroscientist 20, 150–159. doi: 10.1177/1073858413494269 Ziadni, M. S., Sturgeon, J. A., and Darnall, B. D. (2018). The relationship between Wang, J., Zuo, X., and He, Y. (2010). Graph-based network analysis of resting-state negative metacognitive thoughts, pain catastrophizing and adjustment to chronic pain. functional MRI. Front. Syst. Neurosci. 4:16. doi: 10.3389/fnsys.2010.00016 Eur. J. Pain 22, 756–762. doi: 10.1002/ejp.1160 Frontiers in Neuroscience 16 frontiersin.org