{"paper_id":"4534b2b4-d5e7-4688-9763-0484e7a76852","body_text":"Aalborg Universitet\nPain catastrophizing is associated with pain thresholds for heat, cold and pressure in\nwomen with chronic pelvic pain\nGrundström, H.; Larsson, B.; Arendt-Nielsen, L.; Gerdle, B.; Kjølhede, P.\nPublished in:\nScandinavian Journal of Pain\nDOI (link to publication from Publisher):\n10.1515/sjpain-2020-0015\nCreative Commons License\nCC BY-NC 4.0\nPublication date:\n2020\nDocument Version\nPublisher's PDF, also known as Version of record\nLink to publication from Aalborg University\nCitation for published version (APA):\nGrundström, H., Larsson, B., Arendt-Nielsen, L., Gerdle, B., & Kjølhede, P. (2020). Pain catastrophizing is\nassociated with pain thresholds for heat, cold and pressure in women with chronic pelvic pain. Scandinavian\nJournal of Pain, 20(3), 635-646. https://doi.org/10.1515/sjpain-2020-0015\nGeneral rights\nCopyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright owners\nand it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights.\n            - Users may download and print one copy of any publication from the public portal for the purpose of private study or research.\n            - You may not further distribute the material or use it for any profit-making activity or commercial gain\n            - You may freely distribute the URL identifying the publication in the public portal -\nTake down policy\nIf you believe that this document breaches copyright please contact us at vbn@aub.aau.dk providing details, and we will remove access to\nthe work immediately and investigate your claim.\nDownloaded from vbn.aau.dk on: June 11, 2026\n\nOriginal experimental\nHanna Grundström*, Britt Larsson, Lars Arendt-Nielsen, Björn Gerdle and Preben Kjølhede\nPain catastrophizing is associated with pain \nthresholds for heat, cold and pressure in women \nwith chronic pelvic pain\nhttps:/ /doi.org/10.1515/sjpain-2020-0015\nReceived January 16, 2020; revised March 30, 2020; accepted April \n19, 2020\nAbstract\nBackground and aims: Psychological traits such as pain \ncatastrophizing may play a role in the development of \nchronic pelvic pain (CPP). Pain catastrophizing is the ten-\ndency to amplify negative cognitive and emotional pain \nprocesses. The Pain Catastrophizing Scale (PCS) assesses \nelements of pain catastrophizing divided into three sub -\ngroups of factors (rumination, helplessness and mag -\nnification). Previous studies have shown associations \nbetween CPP and increased pain sensitivity, widespread \ngeneralized hyperalgesia, and decreased pain thresholds, \nbut the relation between pain catastrophizing and specific \npain thresholds has not yet been widely examined in this \npatient group. The aims of this study were (a) to determine \nif catastrophizing is increased in women with CPP com-\npared with pain-free women, (b) to assess the importance \nof pain catastrophizing, psychological distress variables, \nand subjective pain sensitivity for pain thresholds of heat, \ncold and pressure in these two groups, and (c) to deter -\nmine whether psychological variables or pain thresholds \nbest contribute to the differentiation between CPP and \ncontrols.\nMethods: Thirty-seven women with chronic pelvic pain \nwho underwent diagnostic laparoscopy on the suspicion \nof endometriosis participated along with 55  healthy and \npain-free controls. All underwent quantitative sensory \ntesting on six locations on the body to determine heat \n(HPT), cold (CPT) and pressure (PPT) pain thresholds. \nThe PCS, the Pain Sensitivity Questionnaire (PSQ), the \nHospital Anxiety Depression Scale, (HADS) demographics \nand clinical data were collected prospectively. Principal \ncomponent analysis and orthogonal partial least square \nregressions were used to assess the associations between \nPCS scores and pain thresholds.\nResults: The women with CPP scored significantly higher \non PCS than the healthy controls. PCS-helplessness, PCS-\nrumination and HADS-depression were significantly \nassociated with pain thresholds for the whole group. In \nthe CPP group, PCS-rumination, body mass index and \nPSQ were significant regressors for HPT and CPT. The PCS \nand the HADS subscales were strongly intercorrelated in \nwomen with CPP and were stronger regressors of group \nmembership than the three pain thresholds. In the group \nof healthy control women, no relationships were found \nto be significant. The psychological variables were some-\nwhat stronger significant regressors than pain thresholds \n(also significant) for group membership.\nConclusions: Women with CPP have significantly higher \npain catastrophizing scores than women without CPP . The \npain catastrophizing rumination factor is significantly \nassociated with pain thresholds of heat and cold in CPP \nwomen. PCS and HADS are strongly intercorrelated and \nPSQ correlates positively with these variables. It seems \nthat the psychological variables are important for group \ndifferentiation.\nImplications: The results clearly indicate the need for \na multimodal assessment (bio-psycho-social) of CPP \npatients including psychological symptoms such as cata-\nstrophizing, anxiety and depression. The registration of \nsemi-objective pain thresholds captures both specific pain \n*Corresponding author: Hanna Grundström, Department of \nObstetrics and Gynaecology in Norrköping, and Department of \nBiomedical and Clinical Sciences, Linköping University, Norrköping, \nLinköping, Sweden; and Department of Health, Medicine and Caring \nSciences, Linköping University, Linköping, Sweden,  \nE-mail: hanna.grundstrom@liu.se\nBritt Larsson and Björn Gerdle: Pain and Rehabilitation Centre, and \nDepartment of Health, Medicine and Caring Sciences, Linköping \nUniversity, Linköping, Sweden\nLars Arendt-Nielsen: Center for Sensory-Motor Interactions, \nDepartment of Health Science and Technology, Faculty of Medicine, \nAalborg University, Aalborg, Denmark\nPreben Kjølhede: Department of Obstetrics and Gynaecology in \nLinköping, and Department of Biomedical and Clinical Sciences, \nLinköping University, Linköping, Sweden\n© 2020 Scandinavian Association for the Study of Pain. Published by Walter de Gruyter GmbH, Berlin/Boston. All rights reserved.\nSc\nand J Pain 2020; 20(3): 635–646\nPublished online May 09, 2020\n\nsensitivity information (mechanical pressure, cold or heat) \nand the degree of wide spread pain hypersensitivity. There \nis a need for future larger studies investigating whether \ncertain profiles in the clinical presentations (including \npain thresholds and psychological variables) are associ-\nated with outcomes after different types of interventions.\nKeywords: chronic pelvic pain; catastrophizing; PCS; pain \nthresholds.\n1   Introduction\nChronic pelvic pain (CPP) is common in women of repro -\nductive age [1]. It often originates from multifactorial \nmechanisms and can manifest as dysmenorrhoea or endo-\nmetriosis [2]. Chronic pain can lead to widespread reduced \npain thresholds, i.e. generalized pain hypersensitivity. By \ndefinition, pain hypersensitivity as presented by reduced \npain thresholds represents an abnormal state of respon-\nsiveness in the nociceptive pain system [3]. Pain hypersen-\nsitivity can be assessed with Quantitative Sensory Testing \n(QST), a set of psychophysical tools to assess mechanisti-\ncally the status of the nociceptive and non-nociceptive \npathways. The tests commonly include cold, heat and pres-\nsure thresholds, pain detection thresholds or pain toler -\nance thresholds [4]. QST has been used to investigate pain \nsensitivity in women with CPP and has shown increased \nand/or widespread hyperalgesia, elevated sensory thresh-\nolds and myofascial trigger points [5–10]. These results \nmay indicate that CPP may be defined as nociplastic pain, \nthe recently defined new type of pain mechanism defined \nby the International Association of the Study of Pain [11].\nThe perception of pain is integrated with psychologi-\ncal traits and pain coping behaviour and interacts with \nQST assessments [12]. Increased anxiety, depression and \nperceived pain sensitivity correlate with reduced pain \nthresholds [13–16]. Emotional dysregulation has been \nimplicated as a transdiagnostic risk factor for pain sever -\nity and mood aspects [17 , 18]. Emotion regulation, e.g. cat-\nastrophizing, is a trans-diagnostic process that ties pain \nand depression/emotion [19–21]. Pain catastrophizing is \nthe tendency to amplify negative cognitive and emotional \nprocesses related to pain. It influences the sensation of \npain [12] and may be a risk factor for the development of \nchronic pain after surgery [22]. Pain catastrophizing may \nact as a predictor of chronic pain among women with CPP \n[23–25] or endometriosis [26, 27] and in young women \nwith menstrual pain [28]. Moreover, catastrophizing is \none of several contributors to the severity of CPP [29] \nand to reduced quality of life [24]. CPP patients with high \ncatastrophizing showed worse outcomes after different \ntreatments compared with those with low catastrophiz -\ning [25].\nThe relationship between pain catastrophizing and \npain thresholds in various chronic pain conditions has not \nbeen definitively established [30]. A deeper understand -\ning of how pain mechanisms are related to psychological \nfactors is essential for the planning and management of \neffective care for women with CPP . We have reported that \nwomen with CPP had alterations in pain thresholds indi-\ncating widespread hypersensitivity, and a time-dependent \ncorrelation between pain thresholds and duration of the \nCPP . In addition, the pain thresholds were significantly \ncorrelated with pain sensitivity [10, 16].\nThe aims of this study were (a) to determine if cata-\nstrophizing is increased in women with CPP compared \nwith pain-free women, (b) to assess the importance of pain \ncatastrophizing, psychological distress variables, and sub-\njective pain sensitivity for pain thresholds of heat, cold and \npressure in these two groups, and (c) to determine whether \npsychological variables or pain thresholds best contribute \nto the differentiation between CPP and controls.\n2   Methods\n2.1   Study design and sample\nThis is a secondary analysis of the data from a cross-\nsectional observational comparative study that was con-\nducted between December 2013 and June 2016 at the \nDepartment of Obstetrics and Gynaecology at a central \nhospital and a university hospital in southeast Sweden \n[10]. The study was approved by the Regional Ethics Board \nof Linköping University (Reg.no. 2013/19-3).\nIn summary, pain thresholds for heat, cold and pres -\nsure were prospectively measured in 37  women with CPP \nreferred for diagnostic laparoscopy due to symptoms that \ncould indicate endometriosis, and in 55  healthy women \nwithout CPP . All participants filled in the PCS form, the Pain \nSensitivity Questionnaire (PSQ) and the Hospital Anxiety \nand Depression Scale (HADS) form. A detailed description \nof the study is presented in the original study [10].\n2.2   Measurements\n2.2.1   Pain Catastrophizing Scale (PCS)\nThe PCS is a self-administered questionnaire consist -\ning of 13 items divided into three domains: helplessness, \nmagnification, and rumination. The questions cover the \n636 Grundström et al.: Pain catastrophizing is associated with pain thresholds\n\n\npatient’s thoughts and feelings in different painful situ -\nations and include information on both intensity and fre-\nquency. The items are answered on a five point Likert-type \nscale using the phrases: not at all (0), to a slight degree (1), \nto a moderate degree (2), to a great degree (3), and all the \ntime (4). The scores for the three domains are summarised \nby different items: magnification relates to items 6, 7 and \n13, rumination to items 8–11, and helplessness to items 1–5 \nand 12 [31]. Higher scores indicate a greater tendency for \ncatastrophizing. The Swedish version of the PCS has been \nvalidated [32].\n2.2.2  \n Pain Sensitivity Questionnaire (PSQ)\nIn the PSQ [14] the patient grades the imagined painful -\nness of 14 described painful everyday situations, where \nthe situations include different pain types such as blunt, \ncold, hot and sharp, and are divided into different inten-\nsities and body sites. Three items serve as non-painful \nsensory references. The average rating is the PSQ total \nscore, which is calculated from all painful items. A higher \nPSQ score indicates higher perceived pain sensitivity [14].\n2.2.3  \n Hospital Anxiety and Depression Scale (HADS)\nThe HADS is a self-rating test assessing symptoms of \nanxiety and depression using 14 items divided into two, \nseven items subscales (anxiety subscale and depression \nsubscale) [33]. The subscale scores range between 0 and \n21 points, with higher scores indicating more symptoms \nof anxiety or depression. The Swedish version has been \nvalidated [34].\n2.2.4  \n Quantitative Sensory Testing (QST)\nPain thresholds for heat, cold and pressure were meas -\nured on six body sites using QST [35] according to the \nguidelines proposed by the German Research Network on \nNeuropathic Pain, with minor deviations [36]. The body \nsites were: the abdominal wall, seven cm lateral to the \numbilicus on both sides, the abdominal wall just above \nthe symphysis pubis, five cm lateral to the midline on both \nsides, the medial plane of the low back just below the fifth \nlumbar vertebra, and on the dominant leg, four cm dis -\ntally from the tuberositas tibiae (the control area).\nHeat (HCP) and cold (CPT) pain thresholds were \nmeasured with The Medoc TSA II NeuroSensory Analyzer \n(Medoc Ltd. 1 Ha’dekel St. Ramat Yishai 30095 Israel) \nusing a 3 × 3  cm\n2 computer-controlled thermode. The \ntemperature dropped or increased from 32 °C at a rate of  \n1.5 oC/s to a maximum of 50 °C, or a minimum of 0 °C. The \nparticipant stopped the stimulation when first detecting a \npainful stimulus.\nPressure pain threshold (PPT) measurements were \nperformed using a hand-held electronic algometer \n(Sometic AB, Hornby, Sweden) with a pressure surface \narea of 1  cm\n2. Pressure was applied at a rate of approxi-\nmately 40 kPa/s and was discontinued on the first sensa-\ntion of pain.\nThree measurements of each stimulus were performed \non each body site and the arithmetic average was pre-\nsented as the pain threshold [35]. The testing order of both \nbody sites and stimuli was altered randomly. The majority \nof the measurements were performed by the first author, \nand the rest by three research nurses experienced in QST.\n2.3   Statistical analyses\nStatistical analyses were conducted with the software Sta-\ntistica v 13.1 (Dell Software, 5 Polaris Way, Aliso Viejo, CA \n92656, USA) and the SIMCA-P software version 15 (Umet -\nrics, Sartorius Stedim Biotech, Umeå, Sweden). Data are \npresented as median and interquartile range (IQR) or fre-\nquency (number and percent).\nA between-group comparison of demographic char -\nacteristics, clinical characteristics, pain thresholds and \nquestionnaire data was conducted using a Mann-Whitney \nU-test for continuous data and Pearson’s chi-squared test \nor Fisher’s exact test for nominal data. The level of statis-\ntical significance was set at p  < 0.05 for two-sided tests.\nDue to the risk of downplaying the interrelation-\nship among factors and thus reaching incorrect conclu -\nsions when using classic statistical methods (for instance \nlinear regression), and the obvious risk of multicollin-\nearity problems when using psychological variables, \nwe used advanced multivariate data analysis (MVDA) as \nelaborated more in detail elsewhere [16]. The MVDA in \nthe present study consists of principal component analy -\nsis (PCA) to detect outliers, and orthogonal partial least \nsquare regressions (OPLS) for the multivariate regressions. \nOPLS discriminant analysis (OPLS-DA) is used to show \nwhich variables have the largest discriminatory power for \ngroup separation (i.e. CPP vs. controls). Methods such as \nmultiple linear regression assumes that the independent \nvariables (x-variables, regressors) are not strongly inter -\ncorrelated i.e. multicollinearity is not present. Based on \nprevious research we had good reasons to suspect the \npresence of multicollinearity. OPLS is instead based upon \nGrundström et al.: Pain catastrophizing is associated with pain thresholds 637\n\nthe assumption that the independent variables may be \nintercorrelated (in unknown ways) and takes advantage of \nthis multicollinearity pattern.\nThe data is not required to be normally distributed \nwhen applying these methods of MVDA [37]. Basically, \nthe MVDA R\n2 describes the goodness of fit – the fraction \nof sum of squares of all the variables explained by a prin-\ncipal component. Q\n2 describes the goodness of prediction \n– the fraction of the total variation of the variables that \ncan be predicted by a principal component using cross-\nvalidation methods [38].\nA check for outliers was conducted using score plots \nof the PCA in combination with Hotelling’s T\n2, and dis -\ntance to model in X-space [38]. No extreme outliers were \ndetected in the present study. PCA can be regarded as a \nmultivariate correlation analysis. Instead of performing \nmultiple bi-variate correlations a PCA analysis is per -\nformed and the risks associated with multiple testing \nis markedly reduced. Besides checking for multivari-\nate outliers PCA ’s were made in order to understand the \ncorrelation pattern among the variables used as regres -\nsors (x-variables). Graphic presentations implemented in \nSIMCA-P software were used to facilitate the understand -\ning of the correlation pattern.\nOPLS was used to explore the relative roles of PCS sub-\nscales, HADS subscales and PSQ together with background \ndata (age, BMI, and smoking) to explain the variations in \nthe pain thresholds for each stimulus [38]. A variable influ-\nence on projection (VIP) ≥  1.0 was considered significant \nif the VIP value had a 95% jack-knife uncertainty confi-\ndence interval non-equal to zero [38]. P(corr) depicts the \nloading of each variable scaled as a correlation coefficient, \nthus standardising the range from −1 to + 1. An absolute \np(corr) > 0.4–0.5 is generally considered significant [37]. \nVIP values are specific for a certain regression but can be \ncompared within a regression i.e. between the x-variables \n(regressors). P(corr) is suitable for comparisons between \nregressions but require that the same dependent variable \n(Y-variable) is used. For each regression, we report the R\n2, \nQ2, and the result (i.e. p-value) of a cross-validated analysis \nof variance (CV-ANOVA). In the present study we required \na significant CV-ANOVA for a regression to be significant. \nA certain variable was considered a significant variable \nwhen VIP > 1.0 and absolute p(corr) ≥ 0.50.\n3   Results\nThe demographic and descriptive data and the scores of \nthe psychometric instruments for the 37 CPP women and \nthe 55 controls are presented in Table 1. The CPP women \ndeviated significantly from the control women in all \nTable 1: Demographics, clinical characteristics, pain thresholds for heat, cold and pressure, PCS subscales, HADS subscales and PSQ total \nscores of women with chronic pelvic pain and healthy controls.\nVariable   Women with chronic \npelvic pain (n = 37)\n  Control group of \nhealthy women (n = 55)\n p-value\nAge (years)   25.0; 22–30   31.0; 25–35   0.002\nBMI (kg/m2)   23.7; 20.8–26.8   23.9; 21.1–25.7   0.796\nNulli-parous (no. of women)   30 (81)   24 (44)   0.014\nCurrently smoking (no. of women)  9 (24)   2 (4)   0.006\nHormonal birth control \nmedication (no. of women)\n  20 (54)   34 (62)   0.520\nDuration of pelvic pain (months)   36.0; 16–78   –  \nHeat pain threshold (°C)   44.6; 41.0–47.2   47.8; 46.6–49.1  <0.001\nCold pain threshold (°C)   10.5; 6.3–20.5   0.59; 0.0–6.9  <0.001\nPressure pain threshold (kPa)   290.8; 227.3–434.8   553.0;401.5–654.8  <0.001\nPCS rumination   11.0; 10.0–12.8   4.0; 1.8–7.0  <0.001\nPCS magnification   5.5; 4.0–8.0   2.0; 1.0–4.0  <0.001\nPCS helplessness   13.5; 11.0–18.0   3.0; 1.0–5.3  <0.001\nHADS anxiety   10.0; 7.0–12.0   4.0; 2.0–7.0  <0.001\nHADS depression   8.0; 5.0–11.0   2.0; 1.0–3.0  <0.001\nPSQ   4.6; 2.6–4.1   3.6; 3.4–5.3  <0.001\nFurthest to the right is shown the between-group comparisons (p-value).\nFigures denote median; 25th–75th interquartile range or number of women and (%).\nBMI = Body Mass Index; PCS = Pain Catastrophizing Scale; PCS rumination = rumination scale of PCS; PCS magnification = magnification \nscale of PCS; PCS helplessness = helplessness scale of PCS; HADS = Hospital Anxiety and Depression Scale; HADS-Depression = depression \nscale of HADS; HADS-anxiety = anxiety scale of HADS; PSQ = Pain Sensitivity Questionnaire – total index.\n638 Grundström et al.: Pain catastrophizing is associated with pain thresholds\n\n\nvariables accounted for except for BMI and use of hormo -\nnal birth control. The median duration of pelvic pain in the \nCPP women was 36 months (IQR 18–72 months). Likewise, \nand as reported earlier for these two cohorts [10, 16] the \npain thresholds were significantly lower in the CPP group.\n3.1   Catastrophizing – group comparisons\nCatastrophizing scores were significantly higher in the \nCPP women than in the healthy control women according \nto the three aspects (subscales) captured by PCS (Table 1).\n3.2   Regressions of pain thresholds\n3.2.1   Regressions of pain thresholds – all subjects taken \ntogether\nThe associations between the three pain thresholds (y-var-\niables) and the other variables (x-variables) were investi-\ngated in three regressions (Table 2). Highly significant \nregressions (R\n2: 0.30–0.45; all CV-ANOVA p < 0.001) were \nobtained for the three pain thresholds. A mix of variables \nshowed important associations with the three investi-\ngated thresholds. The three analyses presented in Table 2  \nhad in common that PCS-helplessness, PCS-rumination \nand HADS-depression were significantly associated with \nthe pain thresholds.\nThe three scales of PCS and the two scales of HADS \nwere important regressors correlating negatively with \nPPT. Two PCS scales (i.e. helplessness and rumination) \nwere the most important regressors. In summary, psycho-\nlogical distress and catastrophizing were associated with \nlow PPT when analysing all subjects taken together.\nBoth scales of HADS, PSQ and the three PCS scales \nwere positively associated with CPT (Table 2). Depressive \nsymptoms according to HADS and PSQ followed by two \nPCS scales (i.e. helplessness and rumination) were the \nvariables most strongly correlated with CPT.\nFor HPT, a negative association with PSQ, the depres-\nsion subscale of HADS and two of the PCS scales (i.e. help-\nlessness and rumination) were found (Table 2).\n3.2.2  \n Regressions of pain thresholds in women with CPP\nAmong women with CPP , the regression of PPT did not \nreach significance according to the CV-ANOVA (Table 3). \nTwo of the subscales of PCS together with PSQ tended to \nbe most important for PPT.\nTable 2: OPLS regressions of PPT, CPT and HPT in all subjects taken together (n = 92), i.e. women with CPP and healthy control women.\nPPT – all subjects CPT – all subjects HPT – all subjects\nVariables VIP p(corr) Variables VIP p(corr) Variables VIP p(corr)\nPCS-helplessness 1.31  −0.88 HADS-depression 1.26 0.73 PSQ 1.39  −0.79\nPCS-rumination 1.27  −0.85 PSQ 1.25 0.72 PCS-rumination 1.19  −0.67\nPCS-magnification 1.22  −0.82 PCS-rumination 1.16 0.67 HADS-depression 1.14  −0.64\nHADS-anxiety 1.19  −0.79 PCS-helplessness 1.12 0.65 PCS-helplessness 1.11  −0.63\nHADS-depression 1.16  −0.78 HADS-anxiety 1.05 0.60 PCS-magnification 0.98 −0.55\nPSQ 0.91 −0.60 PCS-magnification 1.02 0.59 HADS-anxiety 0.96 −0.54\nAge 0.64 0.43 BMI 0.89 0.52 BMI 0.88 −0.50\nSmoking 0.39 −0.26 Smoking 0.52 0.30 Smoking 0.57 −0.32\nBMI 0.07 −0.05 Age 0.22 −0.12 Age 0.34 0.19\nR2 0.30 R2 0.45 R2 0.43\nQ2 0.27 Q2 0.36 Q2 0.35\nCV-ANOVA p-value <0.001 CV-ANOVA p-value <0.001 CV-ANOVA p-value <0.001\nNumber of subjects (n) 92 Number of subjects (n) 92 Number of subjects (n) 92\nNote that pain duration was not included in the regression since it defines the two groups of subjects.\nVariables with VIP > 1.0 and absolute p(corr) > 0.50 are significant and shown in bold type. The sign of p(corr) indicates the direction of the \ncorrelation with the dependent variable (+ = positive correlation;  −   = negative correlation). The four bottom rows of each regression report \nR\n2, Q2, p-value of the CV-ANOVA and number of subjects (n).\nPPT = mean value of pressure pain thresholds; CPT = mean value of cold pain thresholds; HPT = mean value of heat pain thresholds; \nBMI = body mass index; Smoking = currently smoking (i.e. dummy variable: smoking = 1, non-smoking = 0); HADS = Hospital anxiety and \ndepression scale; HADS-Depression = depression scale of HADS; HADS-anxiety = anxiety scale of HADS; PCS = Pain Catastrophizing Scale; \nPCS-helplessness = helplessness scale of PCS; PCS-rumination = rumination scale of PCS; PCS-magnification = magnification scale of PCS; \nPSQ = Pain Sensitivity questionnaire – total index.\nGrundström et al.: Pain catastrophizing is associated with pain thresholds 639\n\nFor the two significant regressions of the thermal \npain thresholds (R2: 0.51–0.52; CV-ANOVA p-value: 0.029–\n0.049) it was found that the same variables were signifi-\ncant regressors even though their relative importance \ndiffered somewhat (Table 3). Hence, BMI, PSQ and the \nrumination subscale of PCS were the three significant \nregressors; the three variables correlated positively with \nCPT and negatively with HPT. BMI and PSQ were rela-\ntively equally important while the PCS rumination scale \nin both analyses had less importance even though it was \nsignificant.\n3.2.3  \n Regressions of pain thresholds in healthy control \nwomen\nNo significant regressions were found for any of the three \npain thresholds in the healthy control women (data not \nshown).\n3.2.4  \n Intercorrelations among the independent variables \n(the regressors)\nWe also investigated the intercorrelation pattern among \nthe variables used as regressors (x-variables) in the \nregressions above (Tables 2 and 3). Bivariate correlation \nanalyses both for all subjects together and separate for the \ntwo groups clearly indicated intercorrelations between \nthe independent variables (Supplementary Table S1). \nHowever, the interpretations of these are complicated and \ntherefore multivariate correlation analyses by means of \nPCA were performed to investigate whether the regressors \nrepresented one or several groups of variables (latent vari-\nables or components).\nIn all subjects taken together, the PCA resulted in \none significant component (R\n2 = 0.45; Q2 = 0.32) (Fig. 1). \nFigure 1 shows the two HADS scales and the three PCS \nscales as highly positively intercorrelated (they had the \nsame sign and had high p -values according to the first \ncomponent [p1 =  the horizontal axis]). PSQ also had a \nrelatively high loading on p1 but was more distant to \nthe six psychological variables and was less strongly \ncorrelated with them. Smoking, BMI, and age had low \nabsolute loadings (< 0.22) upon p1 and thus were not so \nimportant. Note that the second component p2 (verti-\ncal axis) in Fig. 1 was not significant. In Supplementary \nFigure S1 the three pain thresholds are included in the \nPCA.\nAlso, in the CPP (Fig. 2), PCA resulted in one signifi-\ncant component (R\n2 = 0.31, Q2 = 0.11). The pattern of vari-\nables was very similar to what is shown in Fig. 1. Hence, \nTable 3: OPLS regressions of PPT, CPT and HPT in women with CPP (n = 37).\nPPT – CPP women CPT– CPP women HPT– CPP women\nVariables VIP p(corr) Variables VIP p(corr) Variables VIP p(corr)\nPCS-rumination 1.45  −0.79 BMI 1.89 0.82 PSQ 1.96  −0.83\nPSQ 1.41  −0.75 PSQ 1.76 0.77 BMI 1.83  −0.78\nPCS-magnification 1.38  −0.74 PCS-rumination 1.19 0.52 PCS-rumination 1.01  −0.43\nPCS-helplessness 1.14  −0.61 HADS-depression 0.82 0.36 Pain duration 0.79 0.34\nHADS-depression 1.09  −0.58 Age 0.66 0.29 Age 0.70 −0.30\nHADS-anxiety 0.92 −0.49 PCS-magnification 0.63 0.28 PCS-magnification 0.59 −0.25\nBMI 0.79 −0.42 Pain duration 0.52 −0.23 HADS-depression 0.48 −0.21\nSmoking 0.19 0.10 PCS-helplessness 0.31 0.14 PCS-helplessness 0.23 −0.10\nPain duration 0.14 0.08 HADS-anxiety 0.15 0.07 Smoking 0.17 −0.07\nAge 0.10 −0.05 Smoking 0.09 0.04 HADS-anxiety 0.12 −0.05\nR2 0.17 R2 0.52 R2 0.51\nQ2 0.02 Q2 0.28 Q2 0.23\nCV-ANOVA p-value 0.702 CV-ANOVA p-value 0.029 CV-ANOVA p-value 0.049\nNumber of subjects (n) 37 Number of subjects (n) 37 Number of subjects (n) 37\nVariables with VIP > 1.0 and absolute p(corr) > 0.50 are significant and shown in bold type. The sign of p(corr) indicates the direction of the \ncorrelation with the dependent variable (+ = positive correlation;  −   = negative correlation). The four bottom rows of each regression report \nR\n2, Q2, p-value of the CV-ANOVA and number of subjects (n).\nPPT = mean value of pressure pain thresholds; CPT = mean value of cold pain thresholds; HPT = mean value of heat pain thresholds; \nBMI = body mass index; Smoking = currently smoking (i.e. dummy variable: smoking = 1, non-smoking = 0); HADS = Hospital anxiety and \ndepression scale; HADS-Depression = depression scale of HADS; HADS-anxiety = anxiety scale of HADS; Pain duration = pain duration in \nmonths; PCS = Pain Catastrophizing Scale; PCS-helplessness = helplessness scale of PCS; PCS-rumination = rumination scale of PCS; PCS-\nmagnification = magnification scale of PCS; PSQ = Pain Sensitivity questionnaire – total index.\n640 Grundström et al.: Pain catastrophizing is associated with pain thresholds\n\n\nin CPP the two HADS scales and the three PCS scales were \nhighly intercorrelated (positively, i.e. they had the same \nsign and had high absolute p-values according to p1). \nThe PSQ had a relatively high absolute loading on p1 but \nwas more distant to the six psychological variables (not \nso strongly correlated with them). Smoking, BMI, and age \n–0.5\nPCS_helpless\nPCS_magnif\nHADS_Anx\nPain_dur\nHADS_Depr\nSmoking\nPSQ\nBMI\nAge\nPCS_rumin\n–0.5\n–0.4\n–0.3\n–0.2\n–0.1\n0\n0.1\n0.2\n0.3\n–0.4 –0.3 –0.2 –0.1\np[1]\np[2]\n0 0.1\nFig. 2: PCA of the independent variables (X-variables) used in the regression of pain thresholds (cf. Table 3) in CPP (n = 37). Note that the \nsecond component p2 was not significant (only shown in order to improve interpretation). BMI = body mass index; Smoking = currently \nsmoking (i.e. dummy variable: smoking = 1, non-smoking = 0); HADS-Depr = Hospital Anxiety and Depression Scale -depression scale; \nHADS-Anx = anxiety scale of HADS; Pain dur = pain duration; PCS = Pain Catastrophizing Scale; PCS-helpless = helplessness scale of PCS; \nPCS-rumin = rumination scale of PCS; PCS-magnif = magnification scale of PCS; PSQ = Pain Sensitivity questionnaire – total index.\n–0.5\nPCS_helpless\nPCS_magnif\nHADS_Anx\nHADS_Depr\nSmoking\nPSQ\nBMI\nAge\nPCS_rumin\n–0.6\n–0.4\n–0.2\n0\n0.2\n0.4\n0.6\n–0.4 –0.3 –0.2 –0.1\np[1]\np[2]\n0 0.1 0.2\nFig. 1: PCA of the independent variables (X-variables) used in the regression of pain thresholds (cf. Table 2) in all subjects taken together \n(n = 92). Note that the second component p2 was not significant (only shown in order to improve interpretation). BMI = body mass index; \nSmoking = currently smoking (i.e. dummy variable: smoking = 1, non-smoking = 0); HADS-Depr = Hospital Anxiety and Depression Scale; \n=depression scale of HADS; HADS-Anx = anxiety scale of HADS; PCS = Pain Catastrophizing Scale; PCS-helpless = helplessness scale of PCS; \nPCS-rumin = rumination scale of PCS; PCS-magnif = magnification scale of PCS; PSQ = Pain Sensitivity questionnaire – total index\nGrundström et al.: Pain catastrophizing is associated with pain thresholds 641\n\nhad low absolute loadings (<0.15) upon p1. Note that the \nsecond component p2 is only shown to improve interpreta-\ntion. It was not possible to obtain a significant PCA in the \ncontrol group of healthy women. In Supplementary Figure \nS2 the three pain thresholds are included in the PCA.\nTo summarise both the analysis of all subjects and the \nseparate analysis of the CPP group (Figs. 1 and 2) showed \nthat scales of HADS and PCS were strongly intercorrelated \nand that PSQ correlated positively with these variables. \nHence, we found no evidence that the regressors in Tables \n2 and 3 represented several groups of variables.\n3.3   Relative importance of psychological \nvariables and pain thresholds for group \ndifferentiating\nGroup membership (y-variable) was regressed using \nthe background variables, psychological variables, sub -\njective pain sensitivity (PSQ) and pain thresholds as \nregressors (x-variables). From this significant OPLS-DA \n(Table 4) it can be concluded that the subscales of PCS \nand HADS were somewhat stronger regressors according \nto VIP and p(corr) than the three pain thresholds. PSQ, \nage, smoking and BMI were not significant regressors \nin this context (i.e. VIP <  1.0). Hence, the psychological \nvariables contributed somewhat better than the pain \nthresholds to differentiating between CPP and healthy \npain-free controls.\nThis conclusion was further demonstrated in two \nadditional analyses. When only including the psychologi-\ncal variables including PSQ together with the background \nvariables, a model with higher explained variation was \nachieved compared to when the pain thresholds together \nwith background data and PSQ were used as regressors \nof group membership (R\n2 = 0.62, Q2 = 0.60, CV-ANOVA \np-value < 0.001 versus R 2 = 0.43, Q2 = 0.40, CV-ANOVA  \np-value < 0.001). Hence, these two analyses confirm the \nconclusion that the psychological variables were some-\nwhat more important than pain thresholds for group \ndifferentiation.\n4   Discussion\n4.1   Summary of findings\nImportant results of the present study were:\n – Women with CPP reported more catastrophizing than \nhealthy pain-free women.\n – In all subjects taken together, the three regressions \nof the pain thresholds had in common that PCS-help -\nlessness, PCS-rumination and HADS-depression were \nsignificant regressors.\n – In the group of women with CPP , the rumination sub-\nscale of PCS, BMI, and PSQ were significantly associ-\nated with HPT and CPT.\n – The subscales of HADS and PCS were somewhat \nstronger regressors of group membership (control or \nCPP) than the three pain thresholds.\nTable 4: OPLS-DA of group membership (i.e. healthy control women \ndenoted 0 and women with CPP denoted 1) using psychological \nvariables, pain thresholds and background variables as regressors \n(x-variables).\nVariables VIP p(corr)\nPCS-helplessness 1.27 0.84\nPCS-rumination 1.22 0.81\nPCS-magnification 1.15 0.77\nHADS-anxiety 1.15 0.76\nHADS-depression 1.13 0.75\nCPT 1.12 0.74\nHPT 1.09 −0.72\nPPT 1.07 −0.71\nPSQ 0.92 0.61\nAge 0.57 −0.38\nSmoking 0.45 0.30\nBMI 0.14 0.09\nR2 0.64\nQ2 0.62\nCV-ANOVA p-value >0.001\nNumber of subjects (n) 92\nSignificant variables in bold type. Note that pain duration was \nnot included in the regression since it defines the two groups of \nsubjects.\nVIP (VIP > 1.0 is significant) and p(corr) are reported for \neach regressor. The sign of p(corr) indicates the direction \nof the correlation with the dependent variable (+ = positive \ncorrelation;  −   = negative correlation). The four bottom rows of each \nregression report R\n2, Q2, and p-value of the CV-ANOVA and number \nof subjects included in the regression (n).\nPPT = mean value of pressure pain thresholds; CPT = mean value \nof cold pain thresholds; HPT = mean value of heat pain thresholds; \nBMI = body mass index; Smoking = currently smoking (i.e. \ndummy variable: smoking = 1, non-smoking = 0); HADS = Hospital \nanxiety and depression scale; HADS-Depression = depression \nscale of HADS; HADS-anxiety = anxiety scale of HADS; PCS = Pain \nCatastrophizing Scale; PCS-helplessness = helplessness \nscale of PCS; PCS-rumination = rumination scale of PCS; PCS-\nmagnification = magnification scale of PCS; PSQ = Pain Sensitivity \nquestionnaire – total index.\n642 Grundström et al.: Pain catastrophizing is associated with pain thresholds\n\n\n4.2   Interpretation of results in relation to \ncurrent knowledge/literature\nOur results conform with previous studies proposing high \npain catastrophizing in populations with chronic pain \n[12, 22–24, 26, 27 , 29, 30]. Catastrophizing is a significant \nvariable for quality of life and treatment outcomes, where \nmore catastrophizing leads to lower quality of life and less \npositive treatment outcomes. Furthermore, catastrophiz -\ning is a predictor for poor outcome after surgery [39]. The \nrelationships show the importance of taking catastrophiz-\ning and other psychological aspects into account when \ntreating people with chronic pain [24, 27 , 29]. Also, the \nregression of group membership (Table 4) showed that \ncatastrophizing aspects together with other psychological \naspects were associated with the clinical presentation in \nCPP . However, the relationships between catastrophizing \nand pain thresholds is somewhat different in the literature \nand in the present study.\nThe OPLS regressions of CPT and HPT were signifi-\ncant, while the regression of PPT was not. The reasons for \nthis need further evaluation. Meinits et al. [40] explored \ncatastrophizing as a mediator for pain sensation in \npatients with chronic low back pain and found that PPT \nwas inversely associated with pain catastrophizing [40]. \nAlso, studies of chronic neck pain reported such a nega-\ntive association between PPT and catastrophizing [41, 42]. \nPartly in contrast, Walton et al. [43] analysed the pheno -\ntypes of individuals with neck pain in five international \nregisters and found the association between PPTs and \npain characteristics, including catastrophizing, to be con-\nflicting [43].\nThe psychological impact of the pain experience in \nCPP has been the focus of many studies, of which some \nare synthesised in the recent review by Till et  al. [25]. \nThere is convincing evidence for the theory of pain and \nemotion regulation as a trans-diagnostic process [21], \nwhere anxiety, depression and catastrophizing are highly \ncollinear and together may influence the pain experience, \nand vice versa [25]. Martinez-Calderon et al. [44] recently \nreported that the diagnosis of depression had a stronger \nassociation with pain hypersensitivity than pain catastro-\nphizing in patients with chronic shoulder pain [44]. In our \nresults, the influence of PCS-rumination was significant \nfor the regressions of the two thermal pain thresholds and \na similar tendency was found in the non-significant regres-\nsion of PPT in the CPP group. The fact that PCS-rumination \nwas consistently significant while depression and anxiety \nwere not in the CPP group is interesting and indicates the \nimportant role of catastrophizing in the pain experience \nof CPP . The significance of the rumination scale of PCS \nmay also suggest that rumination should be specifically \ntargeted in the planning of new interventions. McPeak \net al. [27] analysed the PCS subscales in relation to pain \nhealth-related quality of life. They found the strongest \nassociation between high helplessness and poor quality \nof life and suggested treatment strategies to manage pain \ncatastrophizing [27]. The importance of taking a multi-dis-\nciplinary approach and thereby including these types of \npsychological aspects in the treatment and care of women \nwith CPP has frequently been highlighted in the literature \n[19, 23, 25]. This conclusion is further supported by the \nresults of the present study where emotional stress (i.e. \nsymptoms of depression and anxiety) and catastrophiz -\ning aspects were increased in CPP and also constituted \nsignificant regressors for group membership. In order \nto understand the influence of psychological variables \non the development of CPP , it may also be important to \ninclude pain catastrophizing.\nPrevious studies have shown an age-dependent factor \nin pain catastrophizing, where younger age was associ-\nated with higher pain catastrophizing intensity [22, 26]. \nWe failed to demonstrate a significant age-dependent \nimpact on pain catastrophizing. In contrast, BMI was a \nsignificant regressor for the two thermal pain thresholds \nbut not for the pressure threshold. The scanty research \non the influence of BMI upon pain sensitivity in women \nwith CPP is ambiguous. Yosef et al. [29] found a positive \nassociation between high pain sensitivity and high BMI \n[29] while Gurian et al. [45] found higher pain thresholds \namong women with overweight compared with those with \nnormal weight or with obesity [45]. This indicates that \nthe influence of BMI upon pain thresholds merits further \nresearch.\nThe data pinpoints that the psychological variables i.e. \nsymptoms of anxiety and depression and the three aspects \nof catastrophizing together with the three pain thresholds \nsignificantly explained group membership. Hence, both \nthe psychological variables and the pain thresholds con-\ntribute to the understanding of the clinical presentation. \nMoreover, the five psychological variables were – accord -\ning to VIP and p(corr) – somewhat more important than \nthe three pain thresholds even though the three latter \nvariables were also significant. From a clinical perspec -\ntive, this information may be important since registration \nof pain thresholds is associated with short periods of pain \nand, moreover, is also considerably more time-consum-\ning than filling out questionnaires on HADS and PCS. As \nGrundström et al.: Pain catastrophizing is associated with pain thresholds 643\n\nreported by us previously, PSQ is associated with the three \npain thresholds and higher in women with CPP [16] but in \nthe present study with a relatively comprehensive set up of \nvariables, it can be concluded that PSQ was not significant \nin the regression (OPLS-DA) of group membership. PSQ \ncaptures the overall perceived pain sensitivity while the \npain thresholds – semi-objective measures – capture both \nstimuli specific information (mechanical pressure, cold or \nheat) and the degree of spatial spreading [10].\n4.3   Strengths and limitations\nA strength of this study was the use of validated instruments \nand methods that provided information on different aspects \nof the subject. All women with pain had experienced pelvic \npain for at least 4  months, a period which is often set as \nthe minimum for considering the pain as chronic [46]. \nAnother strength was the use of MVDA, which is designed \nto use the properties of the data set in an optimal way. This \napproach differs from classical statistical methods, such \nas multiple linear regression, which have a tendency to \nquantify the level of relations of individual factors and at \nthe same time disregard interrelationships among differ -\nent factors [47]. We also avoided the use of multiple linear \nregression due to the risk of multicollinearity problems in \nour data set. Although parity differed significantly between \nthe groups it was not possible to include this variable in the \nMVDA since it did not fulfil the predetermined criteria (i.e. \nit had a VIP value with a 95% jack-knife uncertainty confi-\ndence interval including zero). Even though the number of \nparticipants in the study was sufficient to enable PCA and \nOPLS, the sample size may be considered small and thus \nmay be a limitation of the study. Due to the very few studies \npublished on the subject and consequently the lack of reli-\nable information in the literature, no power analysis with \nsample size estimation was performed when planning the \nstudy. The sample size was intended to be at least the same \nas the studies previously published in the field. Another \nlimitation may be that the women filled in the PCS after the \nQST, which may have influenced their scoring, even though \nthe results of the QST were not disclosed. The cross-sec -\ntional design of the study and thus the direction of causal-\nity between catastrophizing and pain thresholds cannot be \nanalysed. The relationship may be bidirectional.\n5   Conclusions\nThis study showed that women with CPP , when measured \nby PCS, were more prone to exhibiting catastrophizing \ncompared with healthy pain-free women. This was par -\nticularly evident in regard to the PCS subscale ‘rumina-\ntion’ in the women with CPP . Although the pain thresholds \n(PPT, CPT and HPT) were all significantly associated with \nthe PCS and HADS subscales they were weaker regressors \nof group membership. This underlines the importance of \ntaking patients’ psychological status and coping strate-\ngies, such as catastrophizing, into consideration when \nanalysing the occurrence of pain hypersensitivity as a \nproxy for nociplastic pain in women with CPP .\nAcknowledgments: The authors would like to thank all \nthe participating women, and the research nurses for \npractical assistance in conducting the study.\nAuthors’ statements\nResearch funding: The study was supported financially \nwith grants from the Medical Research Council of South-\neast Sweden, the Swedish Research Council, the County \ncouncil of Östergötland, and Linköping University. L. \nArendt-Nielsen was supported by IMI Paincare.\nConflict of interest: The authors state no conflict of \ninterest.\nInformed consent: Informed consent has been obtained \nfrom all participants included in this study.\nEthical approval: The research related to human use com-\nplies with all the relevant national regulations, institu -\ntional policies and was performed in accordance with the \ntenets of the Helsinki Declaration, and has been approved \nby the Regional Ethics Board of Linkoping\nUniversity (Reg. no. 2013/19-3).\nReferences\n[1] Ayorinde AA, Macfarlane GJ, Saraswat L, Bhattacharya SL. \nChronic pelvic pain in women: an epidemiological perspective. \nWomens Health (Lond Engl) 2015;11:851–64.\n[2]\n Kobayashi H, Yamada Y, Morioka S, Niiro E, Shigemitsu A IF. \nMechanism of pain generation for endometriosis-associated \npelvic pain. Arch Gynecol Obs 2014;289:13–21.\n[3]\n Woolf CJ. Central sensitization: implications for the diagnosis \nand treatment of pain. Pain 2011;152:S2–S15.\n[4]\n Pavlaković G, Petzke F. The role of quantitative sensory testing \nin the evaluation of musculoskeletal pain conditions. Curr Rheu-\nmatol Rep 2010;12:455–61.\n[5]\n Bajaj P , Bajaj P , Madsen H, Arendt-Nielsen L. Endometriosis is \nassociated with central sensitization: a psychophysical con-\ntrolled study. J Pain 2003;4:372–80.\n[6]\n Laursen BS, Bajaj P , Olesen AS, Delmar C, Arendt-Nielsen L. \nHealth related quality of life and quantitative pain measure-\nment in females with chronic non-malignant pain. Eur J Pain \n2005;9:267–75.\n644 Grundström et al.: Pain catastrophizing is associated with pain thresholds\n\n\n[7] He W, Liu X, Zhang Y, Guo SW. Generalized hyperalgesia in \nwomen with endometriosis and its resolution following a suc-\ncessful surgery. Reprod Sci 2010;17:1099–111.\n[8]\n As-Sanie S, Harris RE, Harte SE, Tu FF, Neshewat G, Clauw DJ. \nIncreased pressure pain sensitivity in women with chronic \npelvic pain. Obstet Gynecol 2013;122:1047–55.\n[9]\n Stratton P , Khachikyan I, Sinaii N, Ortiz R, Shah J. Associa-\ntion of chronic pelvic pain and endometriosis with signs of \nsensitization and myofascial pain. Obstet Gynecol 2015;125: \n719–28.\n[10]\n Grundström H, Gerdle B, Alehagen S, Berterö C, Arendt-Nielsen \nL, Kjølhede P . Reduced pain thresholds and signs of sensiti-\nzation in women with persistent pelvic pain and suspected \nendometriosis. Acta Obstet Gynecol Scand 2019;98:327–36.\n[11]\n International Association of the Study of Pain. IASP Terminol-\nogy. Avaiable at: https:/ /www.iasp-pain.org/Education/Con-\ntent.aspx?ItemNumber=1698. Accessed: 26 Dec 2019.\n[12]\n Edwards RR, Dworkin RH, Sullivan MD, Turk DC, Wasan AD. The \nrole of psychosocial processes in the development and mainte-\nnance of chronic pain. J Pain 2016;17:T70–92.\n[13]\n Bär KJ, Brehm S, Boettger MK, Boettger S, Wagner G, Sauer H. \nPain perception in major depression depends on pain modal-\nity. Pain 2005;117:97–103.\n[14]\n Ruscheweyh R1, Marziniak M, Stumpenhorst F, Reinholz J KS. \nPain sensitivity can be assessed by self-rating: development \nand validation of the Pain Sensitivity Questionnaire. Pain \n2009;146:64–74.\n[15]\n Schwier C, Kliem A, Boettger MK, Bär KJ. Increased cold-pain \nthresholds in major depression. J Pain 2010;11:287–90.\n[16]\n Grundström H, Larsson B, Arendt-Nielsen L, Gerdle B, Kjølhede \nP . Associations between pain thresholds for heat, cold and \npressure, and Pain Sensitivity Questionnaire scores in healthy \nwomen and in women with persistent pelvic pain. Eur J Pain \n2019;23:1631–9.\n[17]\n Aldao A, Nolen-Hoeksema S. Specificity of cognitive emotion \nregulation strategies: a transdiagnostic examination. Behav \nRes Ther 2010;48:974–83.\n[18]\n Paulus DJ, Bakhshaie J, Garza M, Ochoa-Perez M, Mayorga NA, \nBogiaizian D, Robles Z, Lu Q, Ditre J, Vowles K, Schmidt NB, \nZvolensky MJ. Pain severity and emotion dysregulation among \nLatinos in a community health care setting: relations to mental \nhealth. Gen Hosp Psychiatry 2016;42:41–8.\n[19]\n Linton SJ, Bergbom S. Understanding the link between depres-\nsion and pain. Scand J pain 2011;2:47–54.\n[20]\n Svanberg M, Stålnacke BM, Enthoven P , Brodda-Jansen G, \nGerdle B, Boersma K. Impact of emotional distress and pain-\nrelated fear on patients with chronic pain: subgroup analysis \nof patients referred to multimodal rehabilitation. J Rehabil Med \n2017;49:354–61.\n[21]\n Linton SJ. A transdiagnostic approach to pain and emotion. J \nAppl Biobehav Res 2013;18:82–103.\n[22]\n Leung L. Pain catastrophizing: an updated review. Indian J \nPsychol Med 2012;34:204–17.\n[23]\n Bryant C, Cockburn R, Plante AF, Chia A. The psychological \nprofile of women presenting to a multidisciplinary clinic for \nchronic pelvic pain: high levels of psychological dysfunction \nand implications for practice. J Pain Res 2016;9:1049–56.\n[24]\n Sewell M, Churilov L, Mooney S, Ma T, Maher P , Grover SR. \nChronic pelvic pain – pain catastrophizing, pelvic pain and \nquality of life. Scand J Pain 2018;18:441–8.\n[25]\n Till S, As-Sanie S, Schrepf A. Psychology of chronic pelvic pain. \nClin Obstet Gynecol 2019;62:22–36.\n[26]\n Martin CE, Johnson E, Wechter ME, Leserman J, Zolnoun DA. \nCatastrophizing: a predictor of persistent pain among women \nwith endometriosis at 1 year. Hum Reprod 2011;26:3078–84.\n[27]\n McPeak AE, Allaire C, Williams C, Albert A, Lisonkova S, Yong \nPJ. Pain catastrophizing and pain health-related quality-of-life \nin endometriosis. Clin J Pain 2018;34;349–56.\n[28]\n Payne LA, Rapkin AJ, Lung KC, Seidman LC, Zeltzer LK, Tsao JC. \nPain catastrophizing predicts menstrual pain ratings in adoles-\ncent girls with chronic pain. Pain Med 2016;17:16–24.\n[29]\n Yosef A, Allaire C, Williams C, Ahmed AG, Al-Hussaini T, Abdel-\nlah MS, Wong F, Lisonkova S, Yong PJ. Multifactorial contribu-\ntors to the severity of chronic pelvic pain in women. Am J \nObstet Gynecol 2016;215:760.e1–14.\n[30]\n Rivest K, Côté JN, Dumas JP , Sterling M, De Serres SJ. Relation-\nships between pain thresholds, catastrophizing and gender in \nacute whiplash injury. Man Ther 2010;15:154–9.\n[31]\n Sullivan MJL, Bishop SR, Pivik J. The Pain Catastrophizing \nScale: development and validation: EBSCOhost. Psychol \nAssess 1995;7:524–32.\n[32]\n Karlsson CLK. Swedish validation of Pain Catastrophizing Scale \n[In Swedish: Svensk validering av Pain Catastrophizing Scale \nsamt sambanden mellan smärtkatastrofiering, sömnproblem \noch ångest respektive depression 2018] Avaiable at: http:/ /\nwww.diva-portal.org/smash/record.jsf?pid=diva2%3A1188911\n&dswid=9314. Accessed:23 September 2019.\n[33]\n Zigmond AS, Snaith RP . The hospital anxiety and depression \nscale. Acta Psychiatr Scand 1983;67:361–70.\n[34]\n Lisspers J, Nygren A, Söderman E. Hospital Anxiety and Depres-\nsion Scale (HAD): some psychometric data for a Swedish \nsample. Acta Psychiatr Scand 1997;96:281–6.\n[35]\n Mücke M, Cuhls H, Radbruch L, Baron R, Maier C, Tölle T, Treede \nRD, Rolke R. Quantitative sensory testing (QST). Schmerz \n2016;1–8. DOI: 10.1007/s00482-015-0093-2.\n[36]\n Rolke R, Magerl W, Campbell KA, Schalber C, Caspari S, Birklein \nF, Treede RD. Quantitative sensory testing: a comprehensive \nprotocol for clinical trials. Eur J Pain 2006;10:77–88.\n[37]\n Wheelock ÅM, Wheelock CE. Trials and tribulations of’ omics \ndata analysis: assessing quality of SIMCA-based multivariate \nmodels using examples from pulmonary medicine. Mol Biosyst \n2013;9:2589–96.\n[38]\n Eriksson L, Johansson E, Kettaneh-Wold N, Trygg J, Wikström \nC, Wold S. Multi- and megavariate data analysis; part I and II. \nUmeå: Umetrics AB, 2006.\n[39]\n Khan RS, Ahmed K, Blakeway E, Skapinakis P , Nihoyannop-\noulos L, Macleod K, Sevdalis N, Ashrafian H, Platt M, Darzi A, \nAthanasiou T. Catastrophizing: a predictive factor for postop-\nerative pain. Am J Surg 2011;201:122–31.\n[40]\n Meints SM, Mawla I, Napadow V, Kong J, Gerber J, Chan ST, \nWasan AD, Kaptchuk TJ, McDonnell C, Carriere J, Rosen B, Gol-\nlub RL, Edwards RR. The relationship between catastrophizing \nand altered pain sensitivity in patients with chronic low-back \npain. Pain 2019;160:833–43.\n[41]\n Muñoz-García D, López-de-Uralde-Villanueva I, Beltrán-\nAlacreu H, La Touche R, Fernández-Carnero J. Patients with \nconcomitant chronic neck pain and myofascial pain in \nmasticatory muscles have more widespread pain and distal \nhyperalgesia than patients with only chronic neck pain. Pain \nMed 2017;18:526–37.\nGrundström et al.: Pain catastrophizing is associated with pain thresholds 645\n\n[42] Sá S, Silva AG. Repositioning error, pressure pain threshold, \ncatastrophizing and anxiety in adolescents with chronic idi-\nopathic neck pain. Musculoskelet Sci Pract 2017;30:18–24.\n[43]\n Walton DM, Kwok TS, Mehta S, Loh E, Smith A, Elliott J, Kamper \nSJ, Kasch H, Sterling M. Cluster analysis of an international \npressure pain threshold database identifies 4 meaningful \nsubgroups of adults with mechanical neck pain. Clin J Pain \n2017;33:422–8.\n[44]\n Martinez-Calderon J, Meeus M, Struyf F, Diaz-Cerrillo JL, \nClavero-Cano S, Morales-Asencio JM, Luque-Suarez A. Psy-\nchological factors are associated with local and generalized \npressure pain hypersensitivity, pain intensity, and function in \npeople with chronic shoulder pain: a cross-sectional study. \nMusculoskelet Sci Pract 2019;44:102064.\n[45] Gurian MB, Mitidieri AM, da Silva JB, da Silva AP , Pazin C, Poli-\nNeto OB, Nogueira AA, dos Reis FJ, Rosa-Silva JC. Measurement \nof pain and anthropometric parameters in women with chronic \npelvic pain. J Eval Clin Pract 2015;21:21–7.\n[46]\n International Association for the Study of Pain. Classification \nof chronic pain: descriptions of chronic pain syndromes and \ndefinitions of pain terms. Washington: IASP Press, 1994.\n[47]\n Jansen JJ, Szymańska E, Hoefsloot HC, Jacobs DM, Strassburg \nK, Smilde AK. Between metabolite relationships: an essential \naspect of metabolic change. Metabolomics 2012;8:422–32.\nSupplementary Material: The online version of this article offers \nsupplementary material (https:/ /doi.org/10.1515/sjpain-2020-0015).\n646 Grundström et al.: Pain catastrophizing is associated with pain thresholds","source_license":"CC0","license_restricted":false}