Next Article in Journal
COVID-19-Induced Acute Respiratory Distress Syndrome Treated with Hyperbaric Oxygen: Interim Safety Report from a Randomized Clinical Trial (COVID-19-HBO)
Previous Article in Journal
Efficacy and Safety of New B Cell-Targeted Biologic Agent for the Treatment of Systemic Lupus Erythematosus: A Systematic Review and Meta-Analysis
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
JCM
Volume 12
Issue 14
10.3390/jcm12144849
jcm-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Validity of the Central Sensitization Inventory to Address Human Assumed Central Sensitization: Newly Proposed Clinically Relevant Values and Associations

by
Ingrid Schuttert
1,
André P. Wolff
1,*,
Rita H. R. Schiphorst Preuper
2,
Alec G. G. A. Malmberg
3,
Michiel F. Reneman
2 and
Hans Timmerman
1,*
1
Pain Center, Department of Anaesthesiology, University Medical Center Groningen (UMCG), University of Groningen, 9713 GZ Groningen, The Netherlands
2
Department of Rehabilitation Medicine, University Medical Center Groningen (UMCG), University of Groningen, 9713 GZ Groningen, The Netherlands
3
Department of Obstetrics and Gynaecology, University Medical Center Groningen (UMCG), University of Groningen, 9713 GZ Groningen, The Netherlands
*
Authors to whom correspondence should be addressed.
J. Clin. Med. 2023, 12(14), 4849; https://doi.org/10.3390/jcm12144849
Submission received: 13 June 2023 / Revised: 15 July 2023 / Accepted: 20 July 2023 / Published: 23 July 2023
(This article belongs to the Section Anesthesiology)
Browse Figures
Review Reports Versions Notes

Abstract

:
Central sensitization cannot be directly demonstrated in humans and thus a gold standard is missing. Therefore, we used human assumed central sensitization (HACS) when associated with humans. The central sensitization inventory (CSI) is a screening questionnaire for addressing symptoms that are associated with HACS. This cross-sectional study compared patients with chronic pain and at least one central sensitivity syndrome with healthy, pain-free controls via ROC analyses. Analyses were performed for all participants together and for each sex separately. Regression analyses were performed on patients with chronic pain with and without central sensitivity syndromes. Based on 1730 patients and 250 healthy controls, cutoff values for the CSI for the total group were established at 30 points: women: 33 points; men: 25 points. Univariate and multivariate regression analyses were used to identify possible predictors for the CSI score in 2890 patients with chronic pain. The CSI score is associated with all independent factors and has a low association with pain severity in women and a low association with pain severity, age, and body mass index in men. The newly established CSI cutoff values are lower than in previous studies and different per sex, which might be of clinical relevance in daily practice and importance in research.

1. Introduction

Central sensitization (CS) is defined as the “Increased responsiveness of nociceptive neurons in the central nervous system to their normal or subthreshold afferent input [1]”. CS is suggested to increase the likelihood of pain becoming chronic [2]. The use of the term CS in clinical practice is under debate because it is only demonstrated in animal studies. It is difficult to distinguish between patients with and without CS based on objective pathophysiological mechanisms [3,4]. A gold standard to assess the presence and severity of CS is not available. Therefore, it has been previously suggested that CS should be described as human assumed central sensitization (HACS) if signs associated with CS are present in humans [5].
The central sensitization inventory (CSI) [6] was developed as a screening instrument to address experienced symptoms that are supposed to be associated with HACS [6]. The CSI is translated and validated in more than forty languages [7], among them the CSI-Dutch language version (SI) [8,9]. To address symptoms of HACS in patients with chronic pain, a cutoff point of 40 out of 100 was established [10]. The previously suggested cutoff scores for patients with various chronic pain disorders range between 11/100 [11] and 40/100 [10,12,13,14]. It is hinted that there might be a need for alternative cutoff values that may differ between different language versions/countries and between patient populations [15]. Until now, there is no specific cutoff value established for the CSI-Dutch language version (Dlv).
It is beneficial to address the presence of HACS in patients with chronic pain because it might affect the choice of an appropriate treatment to reduce HACS and to reduce the chronification of pain [16,17].
Women experience more pain and have a higher pain severity compared to men [18,19,20,21]. Sex differences might extend to the presentation and severity of HACS, and thus influence the estimation of the attribution to HACS and treatment effects. The number of central sensitization syndromes (CSSs) has been reported to be different between men and women with chronic musculoskeletal pain disorders [22]. The sex difference was also present when HACS was addressed with a CSI cutoff point of 40/100, but not when using the CSI total score [22]. Correlations between CSI and other factors also varied between studies. Pain catastrophizing correlated good [23,24], moderate [25,26,27] or low [15,28] with the CSI. Several disability questionnaires (SPADI [24], ODI [29], RMDQ [29,30], and NDI [31] correlated good [24,31] or moderately [29,30,31] with the CSI. The CSI correlated good [27,31], moderately [24,29,32,33] or low [23,34] with pain intensity, while age was weakly correlated with the CSI [15,33].
The primary aim of this study was to establish the cutoff values for the CSI-Dlv in patients with chronic pain. The secondary aims were (1) to address differences between groups (patients with chronic pain and healthy volunteers, males and females and the presence or absence of CSS) and (2) to address what factors are associated with the CSI score.

2. Materials and Methods

The Medical Research Ethics Committee of the University Medical Center Groningen, Groningen, The Netherlands, granted an exemption for the patient study because it was based on patients’ standard medical files (METc 2020/284). An exemption was also provided for the inclusion of healthy volunteers because participation was not burdensome (METc 2021/361). The study was registered in the Dutch Trial Register under number NL9241 before the execution of the study.

2.1. Participants

In this cross-sectional study, data were retrospectively collected from consecutive patients with chronic pain referred to the specialized multidisciplinary academic pain center at the University Medical Center Groningen (UMCG), Groningen, The Netherlands, between November 2017 and October 2021.
The patients filled out the questionnaires online and sent them through RoQua (UMCG, Groningen, The Netherlands), a questionnaire program built into the electronic patient file (EPIC systems, Verona, WI, USA) before the first visit. Patients were included when they completed the CSI part A and part B and were aged ≥18 years. Patients of which sex and gender were not equal were excluded. A control group of healthy volunteers aged ≥18 years was recruited between July 2021 and October 2021. The call for participation of healthy volunteers was executed by advertisements on social media, flyers, and online local news sites in the Netherlands. The healthy volunteers were asked to complete the online questionnaire containing demographic variables and the CSI part A and B via REDCap [35]. Exclusion criteria for the healthy volunteers were the experience of any current pain or pain in the previous week, the use of pain medication or undergoing pain treatment, being diagnosed with a CSS (based on the CSI part B) or reporting the use of anti-depressant or anti-epileptic drugs at the moment of completing the questionnaire.

2.2. Measures

Participating patients’ and healthy volunteers’ descriptives consist of sex, age, weight, height, BMI and CSI total score. The patient descriptives also include pain severity, pain catastrophizing, pain disability, number of pain locations, number of reported CSSs, health-related quality of life, pain location(s), and pain disorders.

2.2.1. Central Sensitization Inventory (CSI)

The CSI [6,8] consists of two parts. The CSI is a screening tool for the possible presence of CSS. It is intended be used as an indication for the possible presence of HACS. Part A assesses 25 health-related items supposed to be associated with HACS. Each item is measured on a 5-point Likert scale ranging from 0 (never) to 4 (always), and the total score ranges from 0 to 100 with a previously established cutoff value of 40 or higher to address the possible presence of HACS. The original CSI [6], and the Dlv [8] showed good internal consistency and test–retest reliability. Part B evaluates the presence of seven CS-associated disorders, i.e., tension headaches/migraines, fibromyalgia, irritable bowel syndrome, restless leg syndrome, temporomandibular joint disorder, chronic fatigue syndrome, and multiple chemical sensitivities; and three associated diagnoses, i.e., depression, anxiety/panic attacks, and neck injury [6]. The number of reported CSSs is based on the presence of one or more CSSs in the CSI part B.

2.2.2. Pain Catastrophizing Scale (PCS)

The PCS [36] assesses thoughts, feelings, and cognitions about pain. For each of the 13 statements, the participant is asked to answer on a scale from 0 (totally not) to 4 (always), and the total score ranges from 0 to 52. The higher the score, the more catastrophizing thoughts are present. The original PCS (Cronbach’s alpha = 0.93 [37], test–retest reliability = 0.75 [36]), as well as the PCS-Dlv (Cronbach’s alpha between 0.85 and 0.9138–40, test–retest reliability = 0.7341), show good reliability.

2.2.3. Pain Disability Index (PDI)

The PDI [38,39] is used to measure how aspects of the patient’s life are disrupted by chronic pain. It lists seven life activity categories (family/home responsibilities, recreation, social activity, occupation, sexual behavior, self-care, and life-support activities), which have to be scored by the patient between 0 (no limitations) to 10 (totally limited). The total score ranges from 0 to 70. A higher score reflects a higher interference of pain with daily activities. The original PDI shows good internal consistency (Cronbach’s alpha = 0.87) [40], and the test–retest reliability is good (ICC = 0.91) [41]. The PDI-Dlv has good internal consistency (Cronbach’s alpha = 0.85) [39] and sufficient test–retest reliability (ICC = 0.78) [39] in patients with chronic low back pain.

2.2.4. SF-12 Health Survey (SF-12)

The SF-12 [42,43] is a shorter version of the SF-36, a participant-reported health-related quality of life survey. It consists of eight sections: vitality, physical functioning, bodily pain, general health perceptions, physical role functioning, emotional role functioning, social role functioning, and mental health. The total score ranges from 0 to 100, and a higher score indicates a better health status. The SF-12 showed sufficient and good internal consistency with Cronbach’s alpha 0.77 [44] (Physical subscale) and Cronbach’s alpha 0.80 [44] (Mental subscale) in patients with back pain. Test–retest reliability shows to be sufficient with ICC = 0.79 [45].

2.2.5. Numeric Rating Scale (NRS) for Pain

The NRS is a pain rating scale in which patients and healthy volunteers were asked to rate their current pain, mean pain, and worst pain over seven days (0 = no pain; 10 = maximum pain imaginable). The NRS shows good test–retest reliability (ICC = 0.95) [46]. The average pain score for the past seven days was used to assess the factor pain severity.

2.2.6. Pain Location

Patients’ pain location was based on self-reported questionnaires used in the pain center. The categories that could be chosen were “head”, “neck, shoulders, high back, and arms”, “elbow, wrist, hand”, “lower back”, “hip, knee”, “foot, ankle”, “chest, abdomen”, “pelvic” and “other”. When the pain was present in more than one location, these patients were categorized as having pain in “multiple locations”. The number of pain locations was based on the number of chosen categories.

2.2.7. Pain Disorder

Pain disorder was categorized in ICD-11 codes [47]: chronic primary pain, chronic cancer pain, chronic posttraumatic and post-surgical pain, chronic neuropathic pain, chronic headache and orofacial pain, chronic visceral pain, and chronic musculoskeletal pain [48,49]. Because ICD-10 codes were used in hospital administration systems when patients were included, these were extracted from the patient files and converted to ICD-11 codes using the “10To11MapToOneCategory” file from the World Health Organization [47,50].

2.3. Sample Size Calculation for the Number of Healthy Volunteers

For the healthy volunteers, a sample size table was used [51] to calculate the required sample size. With a confidence level of 95%, a margin of error of 5.0%, and a population size of 10 million people >18 years old, 384 healthy volunteers were needed.

2.4. Data Analysis

Patients were divided into two groups: patients with one or more CSSs (CSS+) and patients without a CSS (CSS−) based on the CSI part B. Descriptive data are presented per group (total group and per sex separately), CSS+/CSS−, and healthy volunteers. Mean ± standard deviation (SD) was applied when the data were normally distributed and a median and interquartile range [IQR] when not normally distributed (examined via Kolmogorov–Smirnov test). Nominal data are shown in the presence with percentages (n (%)).
For the analyses of our primary aim, the establishment of the cutoff points for the CSI, ROC analyses [52] was conducted. The presence of CSS in patients, defined as at least one CSS based on the CSI part B, versus healthy volunteers without CSS [10] was used in the ROC analyses. The area under the curve (AUC), sensitivity, specificity, positive and negative predictive values, and positive and negative likelihood ratios were calculated [53,54]. The optimal cutoff point was determined based on the Youden index [55] for CSS+ vs. healthy volunteers. This index is defined as J = maxc (Se(c) + SP(c) − 1) and ranges from 0 to 1, with 1 being the optimal diagnostic accuracy. J provides a criterion for choosing the “optimal” threshold value (c*), the threshold value for which Se(c) + Sp(c) − 1 is maximized (maxc) [55,56].
For the analyses of the second aims, differences between the groups (patients compared to healthy volunteers, comparison between males and females, and the comparison between patients with or without a CSS) were tested using the independent student’s t-test in normally distributed data and the Wilcoxon signed-rank test when not normally distributed. For the location of pain, the Χ2-test was used to calculate the differences between CSS+ and CSS−. The number of CSS and the type(s) of CSS were assessed with percentages.
Univariate regression and multivariate regression analyses were used to identify possible predictors for the CSI score and to analyze the associations between independent factors (sex, age, number of central sensitivity syndromes, number of pain locations, pain catastrophizing, pain disability, and pain severity) and CSI score. To address factors associated with the CSI score, a multiple linear regression analysis (backwards method) was performed. With ten factors (sex, age, BMI, pain severity, pain catastrophizing, pain disability, number of pain locations, number of reported CSSs, physical component and mental component SF-12), there are enough patients included in this study (>10 cases per factor [57]) for regression analyses. To choose the selection parameter to decide whether an effect should be retained in the model, 0.01 was used for inclusion and 0.05 for exclusion. This is because of a participant per factor ratio of 100 or above and is based on the Akaike information criterion prognostic models [58,59]. When the selection was made, the included factors were used in linear regression analyses (method enter) to create a regression equation with all the predictor variables. The same analysis is also performed for both sexes separately.
The data and metadata will be stored at the repository at the UMCG, ensuring the data’s security and backup. UMCG pursues a FAIR data policy for research conducted in the UMCG. To make the data findable and accessible to others, we included a description in the UMCG data catalog data: https://www.groningendatacatalogus.nl, accessed on 19 July 2023. The data and metadata are available for researchers inside and outside the institute. This catalog is in sync with relevant (inter)national catalogues, such as Biobanking and Biomolecular Resources Research Infrastructure and National Academic Research and Collaborations Information System. A data access committee has been put in place to review requests and assure the accessibility of the data. This access committee can be reached via the corresponding author.

3. Results

Patient inclusion resulted in 2890 patients consisting of 1813 women (62.7%) (1213 women CSS+ (66.9%) and 600 women CSS− (33.1%)) and 1077 men (37.3%) (517 men CSS+ (48.0%) and 560 men CSS− (52.0%)) with chronic pain (Figure 1a). During patient inclusion, 1036 measurements were excluded because CSI part B was not completed by the patient or was not asked for. The healthy volunteers consisted of 157 women (62.8%) and 93 men (37.2%), a total of 250 participants (Figure 1b). Descriptives are presented for the total group of patients (Table 1) and healthy volunteers (Table 2) and are separated between women and men, and CSS+ and CSS−. The patient’s type of pain, based on the ICD-11 of all participating patients, are presented in Figure 2.
Tree diagram of the ICD-11 codes of all reported ICD-11 codes used 1.0% or more. Some patients reported more than 1 ICD-11 code, but only the main complaint is used. N: number; ♀: women; ♂: men. The total number of patients: n = 2890. Excluded from the tree diagram: codes with a prevalence of <1.0% (n = 511 (17.7%)) and not specified (n = 69 (2.4%)).

3.1. Cutoff Scores for the CSI

Cutoff scores for the CSI were established by comparing CSS+ with healthy volunteers. The ROC analysis resulted in an AUC of 0.95 with the highest Youden index (0.78) at a cutoff score of 30 points on the CSI, resulting in a sensitivity of 85% and a specificity of 93%. (Table 3 and Supplement B). For women, the highest Youden index (0.79) resulted in an AUC of 0.96 and a cutoff score of 33 with a sensitivity of 83% and a specificity of 97% (Table 3). In men, the AUC was 0.95, and the highest Youden index was 0.80, resulting in a cutoff score of 25 with a sensitivity of 89% and a specificity of 91% (Table 3). The total group of CSS+ and healthy volunteers resulted in a positive predictive value of 98.8%, a negative predictive value of 47.8%, a positive likelihood ratio of 11.9, and a negative likelihood ratio of 0.16. In women, the analysis resulted in a positive predictive value of 99.5%, a negative predictive value of 41.8%, a positive likelihood ratio of 25.9, and a negative likelihood ratio of 0.18. In men, it was 98.3%, 59.9%, 10.34, and 0.12, respectively.

3.2. Differences between Groups

Significant differences are observed for the CSI score between CSS+ and CSS− and also when examining the sexes separately (Table 1). When comparing the CSS+ and CSS− with healthy volunteers, significant differences are found in almost all descriptives, except for height, when men with a CSS are compared with healthy male volunteers, and the total group of healthy volunteers is compared with all patients without a CSS (Table 2).
Of the 2890 patients, 59.9% (n = 1730) were diagnosed with one or more CSSs based on the CSI part B. The number of CSSs in the patient group ranges from 0 to 8. Women reported more often one or more CSSs (66.9%, 1213 out of 1813) than men (48.0%, 517 out of 1077). With the increase in the number of CSSs, the mean CSI score increases except for the small number of patients with 8 CSSs. The three most frequently diagnosed CSSs were depression (23.4%), migraine or tension headaches (18.7%), and irritable bowel syndrome (16.5%). In women, the three most frequent CSSs diagnosed were depression (25.4%), migraine or tension headaches (23.3%), and fibromyalgia (21.8%). In men, it was depression (20.1%), followed by neck injury (including whiplash) (12.2%), and migraine or tension headaches (11.0%).

3.3. Associated Factors with the CSI Score

Univariate regression showed that all studied factors were significantly associated with the CSI score, for the total patient group, as well as both sexes separately (Table 4).
For the total patient group, multiple backward linear regression analyses showed that all studied factors, i.e., sex, age, BMI, pain severity, PCS score, PDI score, number of pain locations, number of reported CSSs, the physical and mental component of the SF-12, were associated factors for the CSI score (supplement C). For women, all studied factors (as mentioned for the total patient group), excluding sex and pain severity, were associated with the CSI score (supplement C). In men, the associated factors for the CSI score were the same as in women, except for age and BMI (supplement C). A model was created to predict the CSI score with the associated factors, as shown in Table 5. Based on this model, formulas can be created by multiplying factors with the unstandardized beta.
The formula based on the prediction model is for all patients: 69.71 + sex*2.79 − age*0. 07+ BMI*0.10 − pain severity*0.36 + pain catastrophizing*0.16 + pain disability*0.13 + number of pain locations*2.18 − number of reported CSSs*3.24 – physical component SF-12*0.40 − mental component SF-12*0.73.
For women the formula is 79.43 − age*0.10 + BMI*0.13 + pain catastrophizing*0.09 + pain disability*0.10 + number of pain locations*2.24 + number of CSSs*3.21 − physical component SF-12*0.40 − mental component SF-12*0.80.
In men, the formula is 65.38 + pain catastrophizing*0.23 + pain disability*0.12 + number of pain locations*2.05 + the number of reported CSSs*3.23 − physical component SF-12*0.44 − mental component SF-12*0.63. The number of reported CSSs is the most associated factor, followed by sex and the number of pain locations.

4. Discussion

This study resulted in a CSI cutoff value of 30 in the total group, 33 for women, and 25 for men comparing patients with CSS(s) to healthy volunteers.
The secondary aims of this study revealed high discrimination between the CSS+ group and healthy volunteers. Psychometric analyses in the total group and for sexes separately showed a good probability of a positive test in patients with HACS and a negative test in patients without HACS. The analysis of the CSI score showed that all factors were associated with the CSI score for the total group. The associated factors for the CSI score were sex, age, BMI, pain severity, pain catastrophizing, pain disability, number of pain locations, number of reported CSSs and quality of life. In women, pain severity was eliminated as an associated factor. In men, age, BMI and pain severity were eliminated. The relatively low adjusted R2, thus the explained variance, in this study suggest that more and/or other factors might contribute to the CSI score such as pain sensitivity [23,25,34,60,61] and heart rate variability [23,25,34,60,61,62].
In this study, a larger sample of patients and healthy volunteers with a similar age were used compared to previous studies [10,32]. Neblett and colleagues [10] determined the presence of HACS using a list of eleven CSSs based on the seven CSSs stated in the CSI part B but without the three CS-associated syndromes. However, they added four extra diagnoses: myofascial pain syndrome, posttraumatic stress disorder, interstitial cystitis, and complex regional pain syndrome [10]. This list differs from the CSSs from the CSI part B [12] which was used in our and other studies [11,14,32,62,63] and thus the extra disorders with a HACS component were not structurally examined.
In the original study reporting the development and the validity of the CSI by Neblett and colleagues, the cutoff value was based on the specificity of at least 75% combined with the highest sensitivity score [10,32]. In our and other studies [11,13,32], the Youden index was used. The Youden index reflects the intention of maximizing overall correct classification rates and thus minimizing misclassification rates [53,64]. Using the same method as Neblett and colleagues in our study, our cutoff values would be lower than established now, with higher sensitivity scores but lower specificity scores.
In our and previous studies [10,12,63] the presence of at least one CSS was indicative of the presence of HACS to establish cutoff values. In other studies, the presence of three or more CSSs [14] or the presence of musculoskeletal disorders [32] was indicative of the presence of HACS. We used the criteria of at least one CSS because previous research established a strong correlation between the CSI and HACS diagnoses [10]. The presence of a CSS indicates that the central nervous system is involved in this patients’ pain. Furthermore, there are also studies in which QST measures [11,25,65] were found helpful to measure physical sensitivity as an indicator of the presence of HACS. Finally, there is also a study where the presence of pain was used to establish a cutoff value for the CSI [13]. However, in our second analysis, we showed that the association between pain severity and the CSI score is not significant when analyzing the sexes separately, supporting the idea that presence of HACS is not the same as the presence of pain.
In the paper by Neblett and colleagues [10], the cutoff score was based on a comparison with healthy volunteers and resulted in a score of 40 or higher. Another study used a comparison with healthy volunteers to establish a cutoff value of 37 [32]. Most studies established cutoff scores of the CSI without a comparison with the healthy population as the control [11,12,13,14,25,63,65] resulting in, for example, a cutoff score ranging from 17 [11]–36 [25] in patients with knee osteoarthritis. A comparison between CSI outcomes from a healthy population and an individual patient helps to interpret a patient’s outcome [66,67]. This is because it might indicate the severity of HACS or the reduction in the difference between patients and the healthy population. Moreover, it might show treatment success. The most optimal CSI cutoff value is supposed to help discriminate more reliably between patients suffering from chronic pain with HACS and healthy volunteers, and to distinguish between patients suffering from chronic pain with and without HACS.
In previous studies addressing the CSI, uniformity lacks in factors that were correlated with the CSI score. Some studies presented correlated factors [15,23,24,27,31] (sex, beliefs about exercise and pain, pain catastrophizing, pain severity, and widespread pain), but in other studies, these factors were not associated with the CSI score [15,23,24]. For example, in our study, pain catastrophizing is a predictor for the CSI score, which is in line with some studies showing a good correlation [23,24] between the pain catastrophizing scale and the CSI. However, other studies showed a moderate [25,26,27] or low [15,28] correlation.
The second analysis in this study shows that pain severity is not associated with the CSI score when women and men are assessed separately. This suggests that pain intensity does not influence the CSI score and vice versa. The association between pain severity and the CSI score in the total patient group is statistically significant, but this might be caused by the differences between sexes that are canceled out when analyzed as a group.
Importantly, there was a sex difference in the established cutoff values for the CSI score and the associated factors confirming previous research [22,23]. The cutoff value for women is higher than for men. Differences in pain perception between sexes align with other studies [18,19,21,68]. Women perceive more clinical pain disorders [18], more pain severity [18,20], lower pain thresholds [19], greater ability to discriminate different kinds of pain [19], less tolerance of noxious stimuli [19], and experience more visceral pain [21] compared to men. These differences are also shown in immune system cell populations [20].
The CSI-Dlv is used in the Netherlands [8,15] and the Dutch-speaking part of Belgium [8,9]. In our study, the mean for the CSI was 40.2. This is within the range of other studies using the CSI-Dlv, ranging from 36.1 [15] to 43.9 [9] in patients with chronic pain. In the study by Kregel and colleagues [8], the mean CSI score of the healthy volunteers was 21.6 compared to 16.4 in our study. This might be explained by our stricter sample criteria for ensuring our healthy volunteers being indeed healthy. Their exclusion criteria were no chronic pain or long-term pain during the past five years. In our study, participants using pain medication, anti-epileptics, and anti-depressives who experienced pain during the past week and were currently undergoing pain treatment were excluded.

4.1. Limitations and Strengths

To have a good representation of the actual population of patients with chronic pain, we included 2890 patients with pain from different origins. We used strict selection criteria for the healthy volunteers, which resulted in 250 instead of the necessary 384 healthy volunteers.
To establish cutoff values for the CSI-Dlv, we included only Dutch-speaking participants. Country and culture might impact the CSI score [10,69], as shown in a multi-country study [70], and from comparing results of studies performed in different countries [71,72]. Therefore, we have to be careful to extrapolate these results to other languages and countries.
In future research, the inclusion of psychophysical tests such as QST might be recommended to address the presence of HACS in patients besides the use of questionnaires [11,25,34,60,65,73]. A recent systematic review showed a weak or no correlation between the CSI and QST and found a strong correlation between the CSI and psychological measures [68]. A weak or no correlation does not mean that these indicators cannot be combined. The combination of questionnaires and QST might help to address patients with HACS according to a recently proposed HACS grading system [5]. In addition, a recent review also recommended that a combination of features and methods can be used to discriminate between mechanism-based categories of pain [74]. Questionnaires such as Multisensory Amplification Scale might also give some insights into a related construct of generalized central sensitivity and/or risk factor for HACS [73].
Using a cutoff score in the CSI creates an interpretation favoring dichotomous outcomes. A categorical distribution [7,75,76,77] or a continuous scale may give more clinical insights, suggesting that HACS is present to a greater or lesser extent, which might be a better interpretation of a biological phenomenon. In a recent systematic review, it was concluded that a higher CSI score suggests a higher chance of HACS being present, but it does not reflect the increased nociceptive response [68].
Future research may determine if a categorical or continuous distribution is more suitable than a dichotomous outcome and how these severity levels may be established.

4.2. Clinical Implications

The newly proposed cutoff values will better distinguish between patients with and without HACS because of higher sensitivity and specificity that might result in a higher prevalence of HACS. However, the positive and negative predictive values should also be considered. A patient should be evaluated on more indicators for HACS, such as temporal summation and conditioned pain modulation [5], to determine if HACS is present. Especially for patients with a CSI score below the cutoff value because of the low negative predictive value, it may be a false negative.

5. Conclusions

In conclusion, new cutoff values for the CSI are established in our study, being 33 for women, 25 for men, and 30 out of 100 for both sexes together. This study revealed high discrimination between the CSS+ group and healthy volunteers. The analysis of the CSI score showed that all factors (sex, age, BMI, pain severity, pain catastrophizing, pain disability, number of pain locations, number of reported CSSs, and quality of life) were associated with the CSI score for the total group. When sexes were analyzed separately, pain severity was not an associated factor. Sex is an important associated factor for the CSI score and should be taken into account in the assessment of HACS via the CSI.

Supplementary Materials

The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/jcm12144849/s1. Supplement A: Table S1. CSI cutoff scores, comparing CSS+ (N = 1730) and healthy volunteers (N = 250); comparing female CSS+ (N = 1213) and female healthy volunteers (N = 157); comparing male CSS+ (N = 517) and male healthy volunteers (N = 93); Supplement B: Figure S1. ROC curves all patients, women and men CSS+ vs healthy volunteers; Supplement C: Table S2. Multiple backward linear regression analysis on influencing factors of the CSI score, including 1537 patients with chronic pain (898 women and 639 men).

Author Contributions

Conceptualization, I.S., A.P.W. and H.T.; methodology, H.T. and I.S.; formal analysis, I.S.; data curation, I.S. and H.T.; writing—original draft preparation, I.S., H.T. and A.P.W.; writing—review and editing, A.G.G.A.M., R.H.R.S.P. and M.F.R.; visualization, I.S.; supervision, A.P.W. All authors have read and agreed to the published version of the manuscript.

Funding

This research received no external funding.

Institutional Review Board Statement

The study was conducted in accordance with the Declaration of Helsinki and approved by the Institutional Review Board (or Ethics Committee) of University Medical Center Groningen, Groningen, The Netherlands (METc 2020/284, 29 May 2020 and METc 2021/361, 17 June 2021).

Informed Consent Statement

Patient consent was waived due to non-burdensome participation and retrospective patient file study.

Data Availability Statement

The data and metadata will be stored at the repository at the UMCG, ensuring the data’s security and backup. UMCG pursues a FAIR data policy for research conducted in the UMCG. To make the data findable and accessible to others, we included a description in the UMCG data catalog data: https://www.groningendatacatalogus.nl (accessed on 19 July 2023). The data and metadata are available for researchers inside and outside the institute. This catalog is in sync with relevant (inter)national catalogues, such as Biobanking and Biomolecular Resources Research Infrastructure and National Academic Research and Collaborations Information System. A data access committee has been put in place to review requests and assure the accessibility of the data. This access committee can be reached via the corresponding author.

Acknowledgments

We thank Remko Soer, PT, for assistance with the statistics, discussion, and comments that greatly improved the manuscript.

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. Loeser, J.D.; Treede, R.-D. The Kyoto protocol of IASP basic pain Terminology. Pain 2008, 137, 473–477. [Google Scholar] [CrossRef] [PubMed]
  2. Woolf, C.J. Central sensitization: Implications for the diagnosis and treatment of pain. Pain 2011, 152, S2–S15. [Google Scholar] [CrossRef] [PubMed]
  3. Cayrol, T.; Draper-Rodi, J.; Fabre, L.; Pitance, L.; Van Den Broeke, E.N. Stuck in the middle with you: Why a broad-brush approach to defining central sensitization does not help clinicians and patients. J. Orthop. Sports Phys. Ther. 2021, 51, 204–206. [Google Scholar] [CrossRef]
  4. van den Broeke, E.N. Central sensitization and pain hypersensitivity: Some critical considerations. F1000Research 2018, 7, 1325. [Google Scholar] [CrossRef]
  5. Schuttert, I.; Timmerman, H.; Petersen, K.K.; McPhee, M.E.; Arendt-Nielsen, L.; Reneman, M.F.; Wolff, A.P. The Definition, Assessment, and Prevalence of (Human Assumed) Central Sensitisation in Patients with Chronic Low Back Pain: A Systematic Review. J. Clin. Med. 2021, 10, 5931. [Google Scholar] [CrossRef] [PubMed]
  6. Mayer, T.G.; Neblett, R.; Cohen, H.; Howard, K.J.; Choi, Y.H.; Williams, M.J.; Perez, Y.; Gatchel, R.J. The development and psychometric validation of the central sensitization inventory. Pain Pract. 2012, 12, 276–285. [Google Scholar] [CrossRef] [Green Version]
  7. PRIDEDALLAS.COM. Central Sensitisation Inventory. Available online: https://www.pridedallas.com/questionnaires/ (accessed on 25 February 2023).
  8. Kregel, J.; Vuijk, P.J.; Descheemaeker, F.; Keizer, D.; van der Noord, R.; Nijs, J.; Cagnie, B.; Meeus, M.; van Wilgen, P. The Dutch Central Sensitization Inventory (CSI): Factor analysis, discriminative power, and test-retest reliability. Clin. J. Pain 2016, 32, 624–630. [Google Scholar] [CrossRef]
  9. Kregel, J.; Schumacher, C.; Dolphens, M.; Malfliet, A.; Goubert, D.; Lenoir, D.; Cagnie, B.; Meeus, M.; Coppieters, I. Convergent validity of the Dutch central sensitization inventory: Associations with psychophysical pain measures, quality of life, disability, and pain cognitions in patients with chronic spinal pain. Pain Pract. 2018, 18, 777–787. [Google Scholar] [PubMed]
  10. Neblett, R.; Cohen, H.; Choi, Y.; Hartzell, M.M.; Williams, M.; Mayer, T.G.; Gatchel, R.J. The Central Sensitization Inventory (CSI): Establishing clinically significant values for identifying central sensitivity syndromes in an outpatient chronic pain sample. J. Pain 2013, 14, 438–445. [Google Scholar] [CrossRef] [Green Version]
  11. Mibu, A.; Nishigami, T.; Tanaka, K.; Manfuku, M.; Yono, S. Difference in the impact of central sensitization on pain-related symptoms between patients with chronic low back pain and knee osteoarthritis. J. Pain Res. 2019, 12, 1757–1765. [Google Scholar] [CrossRef] [Green Version]
  12. Neblett, R.; Hartzell, M.M.; Cohen, H.; Mayer, T.G.; Williams, M.; Choi, Y.; Gatchel, R.J. Ability of the central sensitization inventory to identify central sensitivity syndromes in an outpatient chronic pain sample. Clin. J. Pain 2015, 31, 323–332. [Google Scholar] [CrossRef] [PubMed]
  13. Falling, C.; Stebbings, S.; Baxter, G.D.; Gearry, R.B.; Mani, R. Criterion validity and discriminatory ability of the central sensitization inventory short form in individuals with inflammatory bowel diseases. Scand. J. Pain 2021, 21, 577–585. [Google Scholar] [CrossRef] [PubMed]
  14. Orr, N.L.; Wahl, K.J.; Lisonek, M.; Joannou, A.; Noga, H.; Albert, A.; Bedaiwy, M.A.; Williams, C.; Allaire, C.; Yong, P.J. Central sensitization inventory in endometriosis. Pain 2022, 163, e234–e245. [Google Scholar] [CrossRef] [PubMed]
  15. van Wilgen, C.P.; Vuijk, P.J.; Kregel, J.; Voogt, L.; Meeus, M.; Descheemaeker, F.; Keizer, D.; Nijs, J. Psychological Distress and Widespread Pain Contribute to the Variance of the Central Sensitization Inventory: A Cross-Sectional Study in Patients with Chronic Pain. Pain Pract. 2018, 18, 239–246. [Google Scholar] [CrossRef]
  16. Nijs, J.; Leysen, L.; Vanlauwe, J.; Logghe, T.; Ickmans, K.; Polli, A.; Malfliet, A.; Coppieters, I.; Huysmans, E. Treatment of central sensitization in patients with chronic pain: Time for change? Expert Opin. Pharmacother. 2019, 20, 1961–1970. [Google Scholar] [CrossRef]
  17. Graven-Nielsen, T.; Arendt-Nielsen, L. Assessment of mechanisms in localized and widespread musculoskeletal pain. Nat. Rev. Rheumatol. 2010, 6, 599–606. [Google Scholar] [CrossRef]
  18. Bartley, E.J.; Fillingim, R.B. Sex differences in pain: A brief review of clinical and experimental findings. Br. J. Anaesth. 2013, 111, 52–58. [Google Scholar] [CrossRef] [Green Version]
  19. Berkley, K.J. Sex differences in pain. Behav. Brain Sci. 1997, 20, 371–380. [Google Scholar]
  20. Sorge, R.E.; Totsch, S.K. Sex differences in pain. J. Neurosci. Res. 2017, 95, 1271–1281. [Google Scholar] [CrossRef]
  21. Wiesenfeld-Hallin, Z. Sex differences in pain perception. Gend. Med. 2005, 2, 137–145. [Google Scholar] [CrossRef]
  22. Roldan-Jimenez, C.; Perez-Cruzado, D.; Neblett, R.; Gatchel, R.; Cuesta-Vargas, A. Central Sensitization in Chronic Musculoskeletal Pain Disorders in Different Populations: A Cross-Sectional Study. Pain Med. 2020, 21, 2958–2963. [Google Scholar] [CrossRef]
  23. Sharma, S.; Jha, J.; Pathak, A.; Neblett, R. Translation, cross-cultural adaptation, and measurement properties of the Nepali version of the central sensitization inventory (CSI). BMC Neurol. 2020, 20, 286. [Google Scholar] [CrossRef]
  24. Walankar, P.P.; Panhale, V.P.; Patil, M.M. Psychosocial factors, disability and quality of life in chronic shoulder pain patients with central sensitization. Health Psychol. Res. 2020, 8, 8874. [Google Scholar] [CrossRef] [PubMed]
  25. Gervais-Hupe, J.; Pollice, J.; Sadi, J.; Carlesso, L.C. Validity of the central sensitization inventory with measures of sensitization in people with knee osteoarthritis. Clin. Rheumatol. 2018, 37, 3125–3132. [Google Scholar] [CrossRef]
  26. Andias, R.; Silva, A.G. Cross-Cultural Adaptation and Psychometric Properties of the European Portuguese Version of the Central Sensitization Inventory in Adolescents With Musculoskeletal Chronic Pain. Pain Pract. 2020, 20, 480–490. [Google Scholar] [CrossRef]
  27. Madi, M.; Hamzeh, H.; Abujaber, S.; Altubasi, I. Cross cultural adaptation, validity, and reliability of Central Sensitization Inventory in Arabic language. Disabil. Rehabil. 2022, 44, 8075–8083. [Google Scholar] [CrossRef]
  28. Klute, M.; Laekeman, M.; Kuss, K.; Petzke, F.; Dieterich, A.; Leha, A.; Neblett, R.; Ehrhardt, S.; Ulma, J.; Schafer, A. Cross-cultural adaptation and validation of the German Central Sensitization Inventory (CSI-GE). BMC Musculoskelet. Disord. 2021, 22, 708. [Google Scholar] [CrossRef] [PubMed]
  29. Chiarotto, A.; Viti, C.; Sulli, A.; Cutolo, M.; Testa, M.; Piscitelli, D. Cross-cultural adaptation and validity of the Italian version of the Central Sensitization Inventory. Musculoskelet. Sci. Pract. 2018, 37, 20–28. [Google Scholar] [CrossRef] [PubMed]
  30. Mikkonen, J.; Luomajoki, H.; Airaksinen, O.; Neblett, R.; Selander, T.; Leinonen, V. Cross-cultural adaptation and validation of the Finnish version of the central sensitization inventory and its relationship with dizziness and postural control. BMC Neurol. 2021, 21, 141. [Google Scholar] [CrossRef]
  31. Ide, K.; Yasuda, T.; Hasegawa, T.; Yamato, Y.; Yoshida, G.; Banno, T.; Arima, H.; Oe, S.; Mihara, Y.; Ushirozako, H.; et al. Evaluation of the Central Sensitization Inventory Score in elderly adults with musculoskeletal examination. Mod. Rheumatol. 2021, 31, 885–889. [Google Scholar] [CrossRef]
  32. Tanaka, K.; Nishigami, T.; Mibu, A.; Manfuku, M.; Yono, S.; Shinohara, Y.; Tanabe, A.; Ono, R. Validation of the Japanese version of the Central Sensitization Inventory in patients with musculoskeletal disorders. PLoS ONE 2017, 12, e0188719. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  33. Ohashi, Y.; Fukushima, K.; Inoue, G.; Uchida, K.; Koyama, T.; Tsuchiya, M.; Uchiyama, K.; Takahira, N.; Takaso, M. Central sensitization inventory scores correlate with pain at rest in patients with hip osteoarthritis: A retrospective study. BMC Musculoskelet. Disord. 2020, 21, 595. [Google Scholar] [CrossRef] [PubMed]
  34. Hendriks, E.; Voogt, L.; Lenoir, D.; Coppieters, I.; Ickmans, K. Convergent validity of the Central Sensitization Inventory in chronic whiplash-associated disorders; associations with quantitative sensory testing, pain intensity, fatigue, and psychosocial factors. Pain Med. 2020, 21, 3401–3412. [Google Scholar] [CrossRef]
  35. Harris, P.A.; Taylor, R.; Minor, B.L.; Elliott, V.; Fernandez, M.; O’Neal, L.; McLeod, L.; Delacqua, G.; Delacqua, F.; Kirby, J.; et al. The REDCap consortium: Building an international community of software platform partners. J. Biomed. Inform. 2019, 95, 103208. [Google Scholar] [CrossRef]
  36. Sullivan, M.J.L.; Bishop, S.R.; Pivik, J. The Pain Catastrophizing Scale: Development and validation. Psychol. Assess. 1995, 7, 524–532. [Google Scholar] [CrossRef]
  37. Osman, A.; Barrios, F.X.; Kopper, B.A.; Hauptmann, W.; Jones, J.; O’Neill, E. Factor structure, reliability, and validity of the Pain Catastrophizing Scale. J. Behav. Med. 1997, 20, 589–605. [Google Scholar] [CrossRef]
  38. Pollard, C.A. Preliminary validity study of the pain disability index. Percept. Mot. Ski. 1984, 59, 974. [Google Scholar] [CrossRef]
  39. Soer, R.; Koke, A.J.A.; Vroomen, P.C.A.J.; Stegeman, P.; Smeets, R.J.E.M.; Coppes, M.H.; Reneman, M.F. Extensive Validation of the Pain Disability Index in 3 Groups of Patients With Musculoskeletal Pain. Spine 2013, 38, E562–E568. [Google Scholar] [CrossRef]
  40. Tait, R.C.; Pollard, C.A.; Margolis, R.B.; Duckro, P.N.; Krause, S.J. The Pain Disability Index—Psychometric and Validity Data. Arch. Phys. Med. Rehabil. 1987, 68, 438–441. [Google Scholar]
  41. Gronblad, M.; Hupli, M.; Wennerstrand, P.; Jarvinen, E.; Lukinmaa, A.; Kouri, J.P.; Karaharju, E.O. Intercorrelation and Test-Retest Reliability of the Pain Disability Index (Pdi) and the Oswestry Disability Questionnaire (Odq) and Their Correlation with Pain Intensity in Low-Back-Pain Patients. Clin. J. Pain 1993, 9, 189–195. [Google Scholar] [CrossRef] [PubMed]
  42. Hays, R.D.; Sherbourne, C.D.; Mazel, R.M. The RAND 36-items Health Survey 1.0. Health Econ. 1993, 2, 10. [Google Scholar] [CrossRef] [PubMed]
  43. van der Zee, K.I.; Sanderman, R. Het Meten van de Algemene Gezondheidstoestand met de RAND-36: Een Handleiding; University of Groningen: Groningen, The Netherlands, 1993. [Google Scholar]
  44. Luo, X.M.; George, M.L.; Kakouras, I.; Edwards, C.L.; Pietrobon, R.; Richardson, W.; Hey, L. Reliability, validity, and responsiveness of the short form 12-item survey (SF-12) in patients with back pain. Spine 2003, 28, 1739–1745. [Google Scholar] [CrossRef]
  45. Ibrahim, A.A.; Akindele, M.O.; Ganiyu, S.O.; Kaka, B.; Abdullahi, B.B.; Sulaiman, S.K.; Fatoye, F. The Hausa 12-item short-form health survey (SF-12): Translation, cross-cultural adaptation and validation in mixed urban and rural Nigerian populations with chronic low back pain. PLoS ONE 2020, 15, e0232223. [Google Scholar] [CrossRef]
  46. Alghadir, A.H.; Anwer, S.; Iqbal, A.; Iqbal, Z.A. Test-retest reliability, validity, and minimum detectable change of visual analog, numerical rating, and verbal rating scales for measurement of osteoarthritic knee pain. J. Pain Res. 2018, 11, 851–856. [Google Scholar] [CrossRef] [Green Version]
  47. World Health Organization. ICD-11; International Classification of Diseases 11th Revision: The global Standard for Diagnostic Health Information. Available online: https://icd.who.int/en (accessed on 25 February 2023).
  48. Treede, R.D.; Rief, W.; Barke, A.; Aziz, Q.; Bennett, M.I.; Benoliel, R.; Cohen, M.; Evers, S.; Finnerup, N.B.; First, M.B.; et al. Chronic pain as a symptom or a disease: The IASP Classification of Chronic Pain for the International Classification of Diseases (ICD-11). Pain 2019, 160, 19–27. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  49. Treede, R.D.; Rief, W.; Barke, A.; Aziz, Q.; Bennett, M.I.; Benoliel, R.; Cohen, M.; Evers, S.; Finnerup, N.B.; First, M.B.; et al. A classification of chronic pain for ICD-11. Pain 2015, 156, 1003–1007. [Google Scholar] [CrossRef] [Green Version]
  50. World Health Organization. ICD 10 to 11 Mapping. Available online: https://icd.who.int/browse11/Downloads/Download?fileName=mapping.zip (accessed on 30 December 2022).
  51. Advisors, T.R.A. Sample Size Table. Available online: https://www.research-advisors.com/tools/SampleSize.htm (accessed on 25 February 2023).
  52. Zweig, M.H.; Campbell, G. Receiver Operating Characteristic (Roc) Plots—A Fundamental Evaluation Tool in Clinical Medicine (Vol 39, Pg 561, 1993). Clin. Chem. 1993, 39, 1589. [Google Scholar] [CrossRef]
  53. Habibzadeh, F.; Habibzadeh, P.; Yadollahie, M. On determining the most appropriate test cut-off value: The case of tests with continuous results. Biochem. Medica 2016, 26, 297–307. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  54. Brenner, H.; Gefeller, O. Variation of sensitivity, specificity, likelihood ratios and predictive values with disease prevalence. Stat. Med. 1997, 16, 981–991. [Google Scholar] [CrossRef]
  55. Fluss, R.; Faraggi, D.; Reiser, B. Estimation of the Youden Index and its associated cutoff point. Biom. J. J. Math. Methods Biosci. 2005, 47, 458–472. [Google Scholar] [CrossRef] [Green Version]
  56. Greiner, M.; Pfeiffer, D.; Smith, R.D. Principles and practical application of the receiver-operating characteristic analysis for diagnostic tests. Prev. Vet. Med. 2000, 45, 23–41. [Google Scholar] [CrossRef]
  57. Field, A. Discovering Statistics Using IBM SPSS Statistics, 5th ed.; Sage Publications Ltd.: London, UK, 2018. [Google Scholar]
  58. Larsen, D.B.; Laursen, M.; Edwards, R.R.; Simonsen, O.; Arendt-Nielsen, L.; Petersen, K.K. The combination of preoperative pain, conditioned pain modulation, and pain catastrophizing predicts postoperative pain 12 Months after total knee arthroplasty. Pain Med. 2021, 22, 1583–1590. [Google Scholar] [CrossRef]
  59. Heinze, G.; Dunkler, D. Five myths about variable selection. Transpl. Int. 2017, 30, 6–10. [Google Scholar] [CrossRef] [PubMed]
  60. Holm, L.A.; Nim, C.G.; Lauridsen, H.H.; Filtenborg, J.B.; O’Neill, S.F. Convergent validity of the central sensitization inventory and experimental testing of pain sensitivity. Scand. J. Pain 2022, 22, 597–613. [Google Scholar] [CrossRef]
  61. Echeita, J.A.; Dekker, R.; Preuper, H.R.S.; Reneman, M.F. Maximal cardiopulmonary exercise test in patients with chronic low back pain: Feasibility, tolerance and relation with central sensitization. An observational study. Disabil. Rehabil. 2022, 44, 6287–6294. [Google Scholar] [CrossRef] [PubMed]
  62. Hassett, A.L.; Radvanski, D.C.; Vaschillo, E.G.; Vaschillo, B.; Sigal, L.H.; Karavidas, M.K.; Buyske, S.; Lehrer, P.M. A pilot study of the efficacy of heart rate variability (HRV) biofeedback in patients with fibromyalgia. Appl. Psychophysiol. Biofeedback 2007, 32, 1–10. [Google Scholar] [CrossRef] [PubMed]
  63. Zafereo, J.; Wang-Price, S.; Kandil, E. Quantitative Sensory Testing Discriminates Central Sensitization Inventory Scores in Participants with Chronic Musculoskeletal Pain: An Exploratory Study. Pain Pract. 2021, 21, 547–556. [Google Scholar] [CrossRef]
  64. Perkins, N.J.; Schisterman, E.F. The inconsistency of “optimal” cutpoints obtained using two criteria based on the receiver operating characteristic curve. Am. J. Epidemiol. 2006, 163, 670–675. [Google Scholar] [CrossRef] [Green Version]
  65. Bezerra, M.C.; Bittencourt, J.V.; Reis, F.J.J.; de Almeida, R.S.; Meziat, N.A.M.; Nogueira, L.A.C. Central Sensitization Inventory is a useless instrument for detection of the impairment of the conditioned pain modulation in patients with chronic musculoskeletal pain. Jt. Bone Spine 2021, 88, 105127. [Google Scholar] [CrossRef] [PubMed]
  66. Kendall, P.C.; Marrs-Garcia, A.; Nath, S.R.; Sheldrick, R.C. Normative comparisons for the evaluation of clinical significance. J. Consult. Clin. Psychol. 1999, 67, 285–299. [Google Scholar] [CrossRef] [PubMed]
  67. Kendall, P.C.; Sheldrick, R.C. Normative data for normative comparisons. J. Consult. Clin. Psychol. 2000, 68, 767–773. [Google Scholar] [CrossRef]
  68. Adams, G.R.; Gandhi, W.; Harrison, R.; van Reekum, C.M.; Wood-Anderson, D.; Gilron, I.; Salomons, T.V. Do “central sensitization” questionnaires reflect measures of nociceptive sensitization or psychological constructs? A systematic review and meta-analyses. Pain 2023, 164, 1222–1239. [Google Scholar] [CrossRef]
  69. Cuesta-Vargas, A.I.; Roldan-Jimenez, C.; Neblett, R.; Gatchel, R.J. Cross-cultural adaptation and validity of the Spanish central sensitization inventory. Springerplus 2016, 5, 1837. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  70. Cuesta-Vargas, A.I.; Neblett, R.; Chiarotto, A.; Kregel, J.; Nijs, J.; van Wilgen, C.P.; Pitance, L.; Knezevic, A.; Gatchel, R.J.; Mayer, T.G.; et al. Dimensionality and Reliability of the Central Sensitization Inventory in a Pooled Multicountry Sample. J. Pain 2018, 19, 317–329. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  71. Akeda, K.; Takegami, N.; Yamada, J.; Fujiwara, T.; Nishimura, A.; Sudo, A. Central Sensitization in Chronic Low Back Pain: A Population-Based Study of a Japanese Mountain Village. J. Pain Res. 2021, 14, 1271–1280. [Google Scholar] [CrossRef]
  72. Falling, C.; Stebbings, S.; Baxter, G.D.; Siegel, C.A.; Gearry, R.B.; Nijs, J.; Mani, R. Symptoms of central sensitization in patients with inflammatory bowel diseases: A case-control study examining the role of musculoskeletal pain and psychological factors. Scand. J. Pain 2021, 21, 283–295. [Google Scholar] [CrossRef]
  73. Wang, D.; Frey-Law, L.A. Multisensory sensitivity differentiates between multiple chronic pain conditions and pain-free individuals. Pain 2023, 164, e91–e102. [Google Scholar] [CrossRef]
  74. Shraim, M.A.; Massé-Alarie, H.; Hodges, P.W. Methods to discriminate between mechanism-based categories of pain experienced in the musculoskeletal system: A systematic review. Pain 2021, 162, 1007–1037. [Google Scholar] [CrossRef]
  75. Cuesta-Vargas, A.I.; Neblett, R.; Nijs, J.; Chiarotto, A.; Kregel, J.; van Wilgen, C.P.; Pitance, L.; Knezevic, A.; Gatchel, R.J.; Mayer, T.G. Establishing Central Sensitization–Related Symptom Severity Subgroups: A Multicountry Study Using the Central Sensitization Inventory. Pain Med. 2020, 21, 2430–2440. [Google Scholar] [CrossRef]
  76. Neblett, R.; Hartzell, M.M.; Mayer, T.G.; Cohen, H.; Gatchel, R.J. Establishing clinically relevant severity levels for the central sensitization inventory. Pain Pract. 2017, 17, 166–175. [Google Scholar] [CrossRef]
  77. van der Noord, R.; Paap, D.; van Wilgen, C.P. Convergent validity and clinically relevant categories for the Dutch Central Sensitization Inventory in patients with chronic pain. J. Appl. Biobehav. Res. 2018, 23, e12119. [Google Scholar] [CrossRef]
Jcm 12 04849 g001a 550Jcm 12 04849 g001b 550
Figure 1. (a). Flowchart showing the inclusion of patients. (b). Flow chart showing the inclusion of healthy volunteers. Abbreviations: N: number; CSI: central sensitization inventory; CSS: central sensitivity syndromes.
Figure 1. (a). Flowchart showing the inclusion of patients. (b). Flow chart showing the inclusion of healthy volunteers. Abbreviations: N: number; CSI: central sensitization inventory; CSS: central sensitivity syndromes.
Jcm 12 04849 g001aJcm 12 04849 g001b
Jcm 12 04849 g002 550
Figure 2. Tree diagram of patients’ type of pain based on ICD-11; Green: ICD-11 code.
Figure 2. Tree diagram of patients’ type of pain based on ICD-11; Green: ICD-11 code.
Jcm 12 04849 g002
Table
Table 1. Descriptives of the patients.
Table 1. Descriptives of the patients.
PatientsTotal GroupWomenMenCSS+CSS−
CSS+CSS− CSS+CSS− CSS+CSS− Women vs. MenWomen vs. Men
NMean ± SD
%		NMean ± SD
%		pNMean ± SD
%		NMean ± SD
%		pNMean ± SD %NMean ± SD %ppp
Age (years)173049.5 ± 15.2116050.8 ± 17.10.028 *121347.9 ± 15.260047.9 ± 17.90.94751753.1 ± 14.556054.1 ± 15.70.326<0.001 *<0.001 *
Height (cm)1370173.2 ± 9.5992175.2 ± 10.1<0.001 *955169.2 ± 7.2508168.7 ± 7.20.287415182.6 ± 7.1484182.0 ± 8.00.210<0.001 *<0.001 *
Weight (kg)137082.4 ± 18.599282.2 ± 17.90.88295578.3 ± 17.150874.8 ± 16.2<0.001 *41591.8 ± 18.348490.0 ± 16.30.135<0.001 *<0.001 *
BMI (kg/cm²)136927.4 ± 5.599226.7 ± 5.10.002 *95427.4 ± 5.850826.3 ± 5.70.001 *41527.5 ± 5.048427.1 ± 4.30.2860.7170.008 *
CSI (0−100)173044.8 ± 14.5116033.3 ± 14.2<0.001 *121346.0 ± 14.460034.6 ± 14.4<0.001 *51742.0 ± 14.356031.9 ± 13.9<0.001 *<0.001 *0.001 *
Pain now (0−10)11476.3 ± 2.29196.1 ± 2.30.026 *7856.3 ± 2.24636.0 ± 2.30.0603626.3 ± 2.24566.1 ± 2.30.1980.8530.651
Mean pain last 7 days (0–10)14306.7 ± 1.810246.7 ± 1.80.6809916.7 ± 1.85236.6 ± 1.80.3824396.7 ± 1.85016.7 ± 1.80.8260.9580.296
Worst pain (0–10)14308.5 ± 1.410248.4 ± 1.50.1559918.5 ± 1.35238.4 ± 1.50.0854398.4 ± 1.55018.4 ± 1.40.9780.2370.667
PCS (0–52)168622.1 ± 11.6114521.8 ± 11.40.445117121.3 ± 11.358620.9 ± 11.40.53351524.0 ± 11.955922.7 ± 11.30.059<0.001 *0.009 *
PDI (0–70)117040.4 ± 13.080038.4 ± 14.70.001 *79740.4 ± 12.938837.9 ± 15.00.005 *37340.5 ± 13.341238.9 ± 14.40.0920.8780.349
Number of pain locations17302.5 ± 2.111601.8 ± 1.3<0.001 *12132.6 ± 2.26001.8 ± 1.3<0.001 *5172.3 ± 1.95601.7 ± 1.3<0.001 *0.001 *0.159
Number of CSSs (0–10)17302.0 ± 1.211600<0.001 *12132.1 ± 1.26000<0.001 *5171.7 ± 1.05600<0.001 *<0.001 *NA
Physical component SF-12 (0–100)158238.9 ± 5.6109039.3 ± 5.90.090110438.7 ± 5.556639.0 ± 5.70.30347839.3 ± 5.952439.6 ± 6.10.4630.0850.154
Mental component SF-12 (0–100)158245.2 ± 7.3109047.4 ± 7.3<0.001 *110445.4 ± 7.156647.5 ± 7.3<0.001 *47844.6 ± 7.852447.2 ± 7.4<0.001 *0.0600.615
Pain locations <0.001 * <0.001 * 0.003 *0.008 *<0.001 *
Head 35 2.0%24 2.1%23 1.9%12 2.0%12 2.3%12 2.1%
Neck, shoulders, high back and arms102 5.9%66 5.7%59 4.9%24 4.0%43 8.3%42 7.5%
Elbow, wrist, hand14 0.8%6 0.5%8 0.7%3 0.5%6 1.2%3 0.5%
Lower back323 18.7%315 27.2%213 17.6%155 25.8%110 21.3%160 28.6%
Hip, knee30 1.7%20 1.7%19 1.6%9 1.5%11 2.1%11 2.0%
Foot, ankle24 1.4%40 3.4%16 1.3%12 2.0%8 1.5%28 5.0%
Chest, abdomen93 5.4%72 6.2%72 5.9%47 7.8%21 4.1%25 4.5%
Pelvis61 3.5%39 3.4%49 4.0%29 4.8%12 2.3%10 1.8%
Other1217.0%1049.0%79 6.5%49 8.2%42 8.1%55 9.8%
Multiple locations92753.6%47440.9%675 55.6%260 43.3%252 48.7%214 38.2%
Abbreviations: CSS+: patients with chronic pain with central sensitivity syndrome(s); CSS−: patients with chronic pain without central sensitivity syndrome(s); N: number; BMI: body mass index; PCS: pain catastrophizing scale; PDI: pain disability index; CSI: central sensitisation inventory; CSS: central sensitivity syndrome(s); NA: Not applicable; SF-12: short-form 12-item health questionnaire; SD: standard deviation. Statistics: Comparison made with a student t-test between women vs. men and CSS+ vs. CSS−. For the location of pain, Χ2-test was used to calculate the differences between groups. * p < 0.05.
Table
Table 2. Descriptives of the healthy volunteers.
Table 2. Descriptives of the healthy volunteers.
Healthy VolunteersTotal GroupWomenMenWomen vs. MenCSS+ vs. Healthy VolunteersCSS− vs. Healthy Volunteers
Total groupWomenMenTotal groupWomenMen
NMean ± SDNMean ± SDNMean ± SDppppppp
Age (years)25043.2 ± 16.115741.6 ± 14.29345.9 ± 18.60.056<0.001 *<0.001 *<0.001 *<0.001 *<0.001 *<0.001 *
Height (cm)250175.8 ± 9.2157170.9 ± 6.293184.0 ± 7.2<0.001 *<0.001 *0.004 *0.0930.4120.001 *0.025 *
Weight (kg)25075.2 ± 14.915769.9 ± 12.79384.2 ± 14.1<0.001 *<0.001 *<0.001 *<0.001 *<0.001 *<0.001 *0.001 *
BMI (kg/cm²)25024.2 ± 3.815723.9 ± 4.09324.8 ± 3.30.067<0.001 *<0.001 *<0.001 *<0.001 *<0.001 *<0.001 *
CSI (0–100)25016.4 ± 8.615717.3 ± 8.79315.1 ± 8.30.049 *<0.001*<0.001 *<0.001 *<0.001 *<0.001 *<0.001 *
Abbreviations: CSS+: patients with chronic pain with central sensitivity syndrome(s); CSS−: patients with chronic pain without central sensitivity syndrome(s); N: number; BMI: body mass index; CSI: central sensitization inventory; SD: standard deviation. Statistics: Comparison made with an unpaired student t-test between women vs. men and CSS+ vs. CSS−. For the location of pain Χ2-test was used to calculate the differences between groups. * p < 0.05.
Table
Table 3. Assessment of the CSI cutoff scores for the total group and per sex, comparing patients with CSS+ and healthy volunteers.
Table 3. Assessment of the CSI cutoff scores for the total group and per sex, comparing patients with CSS+ and healthy volunteers.
Total Group (N = 1730) vs. Healthy Volunteers (N = 250);
AUC: 0.953
Cutoff ScoreYouden IndexSensitivitySpecificity
280.77750.88550.8920
290.77850.87050.9080
300.78180.85380.9280
310.77010.83410.9360
320.76480.82080.9440
Women (N = 1213) vs. Healthy Female Volunteers (N = 157)
AUC: 0.956
Cutoff ScoreYouden IndexSensitivitySpecificity
300.79130.88050.9108
310.78510.86150.9236
320.77820.84830.9299
330.79340.82520.9682
340.77360.80540.9682
350.75050.78240.9682
Men (N = 517) vs. Healthy Male Volunteers (N = 93)
AUC: 0.947
Cutoff ScoreYouden IndexSensitivitySpecificity
230.74970.91100.8387
240.76740.90720.8602
250.80370.88970.9140
260.78830.87430.9140
270.77080.85690.9140
Abbreviations: N: number; CSI: central sensitization inventory; CSS: central sensitivity syndromes; AUC: area under the curve. Statistics: ROC analyses were used to calculate AUC, sensitivity and specificity, and the Youden index.
Table
Table 4. Univariate regression analyses for the total patient group and per sex in relation to the CSI total score.
Table 4. Univariate regression analyses for the total patient group and per sex in relation to the CSI total score.
Total GroupWomenMen
NRadj. R2IβpNRadj. R2IβpNRadj. R2Iβp
Sex28900.1710.02931.3145.450<0.001
Age28900.158 0.02547.803−0.152<0.00118130.1330.01748.266−0.126<0.00110770.1320.01643.746−0.130<0.001
BMI23610.0880.00732.4500.251<0.00114620.1200.01432.8340.318<0.0018990.0430.00132.0850.1340.201
Pain severity24540.2120.04527.3211.784<0.00115140.1920.03630.6091.608<0.0019400.2600.06721.5742.134<0.001
PCS28310.3280.10730.5120.440<0.00117570.3180.10133.2090.429<0.00110740.4060.16424.5700.521<0.001
PDI19700.4010.16022.8760.449<0.00111850.3830.14625.9890.428<0.0017850.4460.19818.1610.481<0.001
Number of Pain locations28900.4000.16032.7303.344<0.00118130.4100.16834.5803.225<0.00110770.3530.12530.2553.262<0.001
Number of CSSs28900.4670.21833.7645.332<0.00118130.4590.21035.1604.935<0.00110770.4290.18332.0275.745<0.001
PC SF-1226720.2150.04662.279−0.574<0.00116700.2050.04163.320−0.558<0.00110020.2190.04757.835−0.545<0.001
MC SF-1226720.5180.26889.414−1.076<0.00116700.5150.26491.992−1.086<0.00110020.5490.30185.461−1.068<0.001
Abbreviations: N: Number of patients; R: Correlation; adj. R2: adjusted determination coefficient (adjusted R squared); I: Intercept; β: unstandardized regression coefficient; p: p-value; BMI: body mass index; PCS: pain catastrophizing scale; PDI: pain disability index; CSSs: central sensitivity syndromes; PC SF12: physical component outcome of the SF-12 health survey; MC SF12: mental component outcome of the SF-12 health survey. Statistics: Simple, univariate, linear regression; statistically significant p-value < 0.05.
Table
Table 5. Models to predict the CSI score for the total group and per sex.
Table 5. Models to predict the CSI score for the total group and per sex.
GroupTotal GroupWomenMen
unadjusted R20.5800.5790.550
adjusted R20.5770.5760.545
βCI 95%βCI 95%βCI 95%
Constant69.7160.88–78.5379.4368.35–90.5265.3853.67–77.08
Sex2.791.72–3.86----
Age(years)−0.07−0.10–−0.03−0.10−0.15–−0.05--
BMI0.100.01–0.200.130.01–0.24--
Pain severity (NRS)−0.36−0.72–−0.01----
Pain catastrophizing (PCS)0.160.11–0.210.090.02–0.160.230.16–0.31
Pain disability (PDI)0.130.08–0.180.100.04–0.160.120.05–0.19
Number of pain locations2.181.91–2.442.241.91–2.562.051.59–2.50
Number of CSSs (CSI part B)3.242.83–3.663.212.71–3.713.232.50–3.96
Physical component (SF-12)−0.40−0.50–−0.30−0.40−0.54–−0.25−0.44−0.60–−0.29
Mental component (SF-12)−0.73−0.82–−0.64−0.80−0.92–−0.67−0.63−0.76–−0.50
Abbreviations: β: unstandardized beta; CI 95%: confidence interval of 95%; BMI: body mass index; NRS: numeric rating scale for pain; PCS: pain catastrophizing scale; PDI: pain disability index; CSSs: central sensitivity syndromes; CSI: central sensitization inventory; SF12: short-form 12-item health questionnaire. Statistics: Multiple regression analysis (backwards).
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Schuttert, I.; Wolff, A.P.; Schiphorst Preuper, R.H.R.; Malmberg, A.G.G.A.; Reneman, M.F.; Timmerman, H. Validity of the Central Sensitization Inventory to Address Human Assumed Central Sensitization: Newly Proposed Clinically Relevant Values and Associations. J. Clin. Med. 2023, 12, 4849. https://doi.org/10.3390/jcm12144849

AMA Style

Schuttert I, Wolff AP, Schiphorst Preuper RHR, Malmberg AGGA, Reneman MF, Timmerman H. Validity of the Central Sensitization Inventory to Address Human Assumed Central Sensitization: Newly Proposed Clinically Relevant Values and Associations. Journal of Clinical Medicine. 2023; 12(14):4849. https://doi.org/10.3390/jcm12144849

Chicago/Turabian Style

Schuttert, Ingrid, André P. Wolff, Rita H. R. Schiphorst Preuper, Alec G. G. A. Malmberg, Michiel F. Reneman, and Hans Timmerman. 2023. "Validity of the Central Sensitization Inventory to Address Human Assumed Central Sensitization: Newly Proposed Clinically Relevant Values and Associations" Journal of Clinical Medicine 12, no. 14: 4849. https://doi.org/10.3390/jcm12144849

APA Style

Schuttert, I., Wolff, A. P., Schiphorst Preuper, R. H. R., Malmberg, A. G. G. A., Reneman, M. F., & Timmerman, H. (2023). Validity of the Central Sensitization Inventory to Address Human Assumed Central Sensitization: Newly Proposed Clinically Relevant Values and Associations. Journal of Clinical Medicine, 12(14), 4849. https://doi.org/10.3390/jcm12144849

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop