Trauma and psychological impact in fibromyalgia and other central sensitization syndromes: the role of anxiety and pain acceptance

Abstract

Recent literature suggests an association between traumatic experiences and central sensitization syndromes (CSS), particularly fibromyalgia (FM). However, few studies have explored the specific variables involved in the relationship between post-traumatic stress disorder (PTSD) and the severity of central sensitization. This study aimed to compare the frequency and characteristics of trauma, PTSD, and clinical symptoms among FM patients, other CSS patients, and healthy controls, and to identify potential mediators between intensity of post-traumatic stress symptoms and central sensitization severity.

A cross-sectional study was conducted with 82 women with FM, 41 women with other CSS, and 43 healthy controls. The assessment protocol included a semi-structured interview, self-report measures of PTSD, central sensitization, pain and pain attitudes, sleep quality, fatigue, anxiety, depression, and disability; and objective measures to assess pressure pain thresholds via a digital algometry and sleep quality via actigraphy. Participants wore the actigraphs for one week. Statistical analyses included descriptive statistics, group comparisons using chi-square tests and one-way ANOVA with post-hoc tests, effect size estimations, correlation analyses, and a multiple mediation model.

FM patients reported higher trauma frequency and severity, and a higher prevalence of PTSD and intensity of post-traumatic stress symptoms than the CSS and healthy participants. The FM group also exhibited greater impairment across all clinical variables assessed, followed by the CSS group and healthy individuals. Within FM patients, those with PTSD showed more severe clinical profile than those without PTSD, highlighting the correlations between the intensity of post-traumatic stress symptoms and several clinical symptoms. In this subgroup of FM and PTSD, anxiety and pain acceptance are significant mediators in the relationship between post-traumatic stress symptoms and central sensitization.

The findings highlight the importance of assessing and addressing PTSD in patients with CSS, particularly FM. Individuals with FM/CSS may require a tailored treatment approach that emphasizes addressing emotions such as anxiety and increasing acceptance. Trauma resolution may have a relevant impact on FM/CSS and contribute substantially to improving the quality of life in this complex patient population.

Fibromyalgia (FM) is a chronic widespread musculoskeletal pain syndrome accompanied by sleep disorders, fatigue, cognitive impairment, and mood disturbances, in addition to various concomitant somatic manifestations [1]. This disorder affects approximately 2–3% of the general population [2] and has a large social and economic impact. The total annual direct cost per patient is estimated to range from $1750 to $35,920 in the US and $1250 to $8,540 in Europe, with medication costs accounting for the largest share of total expenditure [3].

FM belongs to a group of syndromes that present high comorbidity among them and have been collectively called Central Sensitization Syndromes (CSS). CSS or nociplastic pain are an important group of conditions where medical test fails to detect abnormalities whose causality can be proven [4, 5]. CSS include fibromyalgia, chronic fatigue syndrome, multiple chemical sensitivity, irritable bowel syndrome, tension headaches and migraines, and most cases of temporomandibular, pelvic, and back chronic pain [4, 6]. These conditions share, in addition to pain, the presence of sleep disturbances, cognitive-affective alterations, and significant life stress, which has led to the suggestion that it is a common condition with different forms of clinical manifestation [4, 6]. One of the most relevant CSS in terms of its severity, chronicity and the functional limitations it implies is FM [4, 6].

The etiology and pathogenesis of FM and CSS are currently little understood. It has been observed that these conditions share a state of central sensitization, which could be their common pathophysiological basis [4, 6, 7]. Central sensitization is defined as an enhanced response of nociceptive neurons in the central nervous system to normal or subthreshold afferent inputs, resulting in abnormal pain processing [4, 6, 7]. A recent model, called Fibromyalgia Imbalance of Threat and Soothing Systems coherently integrates the available evidence on the neurophysiological and psychosocial aspects involved in FM [8]. This model proposes that FM is based on an imbalance between affect regulation systems, with an overactive threat system and a hypoactive calm system. This imbalance, which may be caused by predisposing biological factors, concomitant pain conditions, traumatic experiences, and learning patterns, generates a continuous source of threat signals to the salience network, also known as the medial cingulo-insular network. Persistent activation of this network, which acts as a multimodal alarm system, results in continued activation of the fight or flight response. It is postulated that this overactivation of the salience network perpetuates the emotional imbalance and biological dysfunction underlying FM symptoms. This model, which would also be applicable to other CSS, highlights bidirectional relationships among the stress response, the state of central sensitization, and symptoms of the FM.

Recent literature suggests a relationship between experiencing physical or psychological traumatic events and the development of chronic pain, particularly FM [9, 10, 11]. For instance, it has been reported that individuals with a history of childhood maltreatment have a 2.72-fold increased risk of developing FM [12]. While the type and timing of trauma may change depending on the specific characteristics of each sample and the assessment instruments used in each study, the severity of post-traumatic stress disorder (PTSD) has been identified as an important mediator between trauma exposure and the development of FM/CSS [11, 13]. The scarce studies exploring the compliance with PTSD diagnostic criteria in FM [14], or other chronic pain conditions [15, 16] clearly show that the impact of trauma is up to four times greater than the trauma itself. However, to date, only a limited number of studies have explored how the presence of PTSD or post-traumatic stress symptoms (PTSS) could relate to FM/CSS symptoms. A systematic review about the relationships between trauma and FM suggests that future research could analyze potential mediators between PTSD/PTSS and FM, such as gender, endocrine abnormalities, autonomic dysregulation and psychopathology (anxiety and depression) [11]. The role of depression has shown conflicting results. A study found that the relationship between PTSS after sexual abuse and pain was mediated by depression [17]; however, some data do not support that the relationship between PTSS and FM is mediated by depression [18, 19]. In contrast, it has been found that anxiety plays a mediating role in the relationship between PTSS and FM status whereas coping styles and sleep quality did not mediate this relationship [20]. Further research is needed to understand the association between PTSD/PTSS and pain, highlighting the preliminary nature of research in this area [8, 10]. Nonetheless, no studies have included cognitive-affective characteristics relevant to the field of chronic pain, such as pain-related attitudes, in addition to the typical symptoms of FM/CSS. In this regard, pain catastrophizing and pain acceptance stand out as particularly relevant variables, as both significantly influence pain intensity, emotional distress and pain-related disability across different clinical populations [21, 22]. In addition, studies have been carried out exclusively with subjective measures, while it may be of interest to add objective measures of pain, such as algometry, or sleep quality, such as actimetry [10, 11].

Exploring the relationship between PTSD/PTSS and FM, as well as identifying potential mediators, is crucial for the advancement in the diagnosis and treatment of these patients. Accordingly, the main objectives of the present study are: First, to compare the frequency and characteristics of trauma, PTSS and clinical characteristics (central sensitization, pain, pain attitudes, sleep quality, fatigue, anxiety, depression, and disability) among FM patients, other CSS patients, and healthy controls. Second, to contrast the clinical profile of FM patients exposed to trauma who develop PTSD with those of FM patients without PTSD. Finally, to identify potential mediators between PTSS intensity and FM status (as assessed by Central Sensitization Inventory). Three main hypotheses were proposed: (1) FM patients will show a higher incidence and severity of trauma, PTSD and PTSS, than other CSS patients and healthy participants; (2) FM patients with PTSD will exhibit a more severe clinical profile than FM patients without PTSD; (3) anxiety and/or pain-related attitudes (pain catastrophizing and pain acceptance) will mediate the relationship between PTSS and FM status.

Method

Participants and procedure

The sample consisted of 82 women with FM (M = 53.20 years, SD = 8.54), 41 women with other CSS (M = 38.16 years, SD = 17.07), and 43 healthy controls (M = 40.71 years, SD = 14.07). Considering that FM and other CSS have a much higher prevalence in women than in men [2], only women were included in the present study. The inclusion criteria for FM and CSS groups were as follows: (a) being a women between 18 and 67 years old, (b) having adequate reading comprehension, (c) for the FM group: having a primary diagnosis of FM according to the criteria of the American College of Rheumatology [1], (d) for the CSS group: having a primary diagnosis of another CSS (excluding FM), specifically chronic fatigue syndrome, irritable bowel syndrome, chronic pelvic pain, temporomandibular joint disorder, and migraine/tension headache, according to the criteria of the International Classification of Diseases (ICD-10) [23], (e) having had the corresponding diagnosis for at least one year prior to the study, and (f) remain on a stable medication regimen for at least one month before the start of the study. Exclusion criteria included: (a) the presence of other rheumatic pain conditions, (b) other significant medical conditions or pregnancy, (c) history of substance or medication abuse, (d) serious psychological disorders other than depression, anxiety, or PTSD (for example, psychotic disorders or severe depression with suicidal ideation), (e) being in active psychological treatment, and (f) being involved in a legal proceeding related to disability. The inclusion criteria for the healthy sample were the following: (a) being a woman between 18 and 67 years old, and (b) having adequate reading comprehension. Exclusion criteria included: (a) having a pain condition, (b) significant medical or psychological illness, or pregnancy, and (c) a history of substance or medication abuse.

Patients with FM were referred from the Rheumatology and Internal Medicine Services at the Virgen de las Nieves University Hospital in Granada, Spain, and from the Fibromyalgia Association of Granada, Spain [removed for blind review]. Rheumatology specialists were responsible for diagnosing FM after a comprehensive medical evaluation, following the 2016 American College of Rheumatology classification criteria [1]. Likewise, the participants with other CSS were selected from clinical contexts through a rigorous clinical interview process, during which they had to provide a diagnosis of CSS made by a medical specialist. Healthy individuals were recruited from community settings (e.g., family members and friends of university students, community associations, etc.), attempting to match them with FM and CSS patients on key sociodemographic variables. A total of 174 patients with FM or CSS and 61 potentially eligible healthy volunteers participated in the initial evaluation, with 123 patients (82 with FM and 41 with CSS) and 43 healthy individuals who met the inclusion criteria being selected as the final sample.

Participants were invited to enroll in a study on health-related attitudes and behaviors. Clinical groups were offered the opportunity to participate in psychological therapy after completing the study. The assessment protocol for all participants involved an initial evaluation session (week 1), during which participants were informed about the study objectives, signed the informed consent form, and then underwent an individual semi-structured interview to gather information on demographic and clinical data, biographical and problem history, current situation, and psychological status. At the end of the session, participants were given a battery of questionnaires to complete at home, assessing physical symptoms, cognitive-affective variables, PTSS, and disability. The PTSS self-report was placed at the end of the battery to avoid potential interference with the other measures, as our experience suggests that recalling significant traumatic events tends to increase emotional distress. During a second session (week 2), the information previously collected was completed, pain thresholds were assessed using algometry, and an actigraphy device was provided for participants to wear over the course of a week to monitor sleep quality. The completed questionnaires were also collected. The actigraphy devices were returned during a third visit (week 3). The study was approved by the Provincial Ethics Committee of Granada (session held on November 25, 2019, minute 11/19) of the Council of Health and Family (Government of Andalusia, Spain).

Measures

The Global Assessment of Post-Traumatic Stress Questionnaire (EGEP-5) assesses PTSD in adults according to the Diagnostic and Statistical Manual of Mental Disorders 5th edition [24, 25]. It is composed of 58 items divided into three sections that refer to the evaluation of traumatic events, symptomatology and the functioning of the individual. The first part includes a list of 11 traumatic events to which the person answers according to the main possibilities of experiencing a trauma that the DSM-5 includes: “I have not experienced it”, “I had a direct experience”, “I witnessed the event” or “It happened to someone close to me” (criterion A). Then, the person had to briefly explain the most important traumatic event, state their age at the time of the event, and how often such event occurred (once, or several occasions, or repeatedly). The second part consists of 28 items that evaluates intrusive symptoms (criterion B), avoidance of stimuli associated with the event (criterion C), negative alterations in cognitions and mood (criterion D), hyperarousal (criterion E), and presence of dissociative symptoms. The third part assesses the impact on daily functioning (criterion G). All items are rated using a 0–3-point scale (“not at all”, “a little”, “quite a lot” and “a lot”). The total symptom score ranges from 0 to 80, with scores from 0 to 21 indicating low intensity, scores between 22 and 56 medium intensity, and scores above 57 indicating high intensity. The EGEP-5 has good internal consistency (Cronbach’s α = 0.91) and satisfactory discriminant (g = 1.27) and convergent validity (r = .78) [25]. In the sample of the present study, Cronbach’s alpha coefficient of the intrusive symptoms was 0.913, of the avoidance of stimuli was 0.887, of cognitive and mood disturbances was 0.904, of hyperarousal and reactivity was 0.878, and of functional impact was 0.845.

The Central Sensitization Inventory (CSI) [26] allows identifying patients with symptoms related to central sensitization. The instrument has two sections. Part A evaluates 25 somatic and emotional health-related symptoms using a Likert scale ranging from 0 “never” to 4 “always”, resulting in a total score ranging from 0 to 100, where 40 is the indicative cut-off point of CSS. Score ranges are categorized as follows: 0–29 (subclinical), 30–39 (mild), 40–49 (moderate), 50–59 (severe), and 60 or higher (extreme sensitization). Part B consists of a list of the 10 diseases most related to CSS and it is not scored. The Spanish version of the CSI has high internal consistency (Cronbach’s α = 0.87) and test-retest reliability [27]. In the sample of the present study, Cronbach’s alpha coefficient of the CSI was 0.958.

The McGill Pain Questionnaire-Short form (MPQ-SF) [28] assesses several pain dimensions using 15 verbal pain descriptors (sensory and affective), with 4 response options (from “no” to “severe”), a current pain index (ranging from 0, “no pain at all”, to 5, “insufferable pain”), and a visual analogue scale (VAS) to evaluate pain intensity during the previous week (from 1 “no pain” to 10 “extreme pain”). The total score for the sensory and affective dimension (ranging from 0 to 45) and the VAS was used in the present study. The Spanish version has appropriate internal consistency (Cronbach’s α = 0.74) and discriminant validity [29]. In the sample of the present study, Cronbach’s alpha coefficient of the sensory affective dimension of MPQ-SF was 0.959.

The Pain Catastrophizing Scale (PCS) [30] consists of 13 items that evaluate three dimensions of catastrophic thinking related to pain: magnification, rumination and helplessness. Items are assessed from 0, “not at all”, to 4 “all the time”, resulting in a total score ranging from 0 to 52 points. The Spanish version of the PCS showed good internal consistency (Cronbach’s α = 0.79), test-retest reliability and sensitivity to change [31]. In the sample of the present study, Cronbach’s alpha coefficient of the PCS was 0.959.

The Chronic Pain Acceptance Questionnaire (CPAQ) [32] is a 20-item self-report designed to assess two dimensions of pain acceptance: activity engagement and pain willingness. Items are rated on a Likert scale from 0 “never true” to 6 “always true”, yielding a total score ranging from 0 to 120. The Spanish version of the CPAQ shown adequate test-retest reliability and internal consistency (Cronbach’s α = 0.83) [33]. In the sample of the present study, Cronbach’s alpha coefficient of the CPAQ was 0.887.

The Pittsburgh Sleep Quality Index (PSQI) [34] assesses sleep quality using 19 items grouped into seven dimensions: subjective sleep quality, sleep latency, sleep duration, habitual sleep efficiency, sleep disturbances, use of sleeping medication, and daytime dysfunction. The total score ranges from 0 to 21, with higher scores indicating greater sleep disturbance and scores > 5 indicating poor sleep quality [34]. The Spanish version of the PSQI shows adequate internal consistency (Cronbach’s α = 0.80), test-retest reliability and convergent validity [35]. In the sample of the present study, Cronbach’s alpha coefficient of the PSQI was 0.888.

The Multidimensional Fatigue Inventory (MFI) [36] is a 20-item self-report that assesses five dimensions of fatigue: general fatigue, physical fatigue, mental fatigue, reduced motivation, and reduced activity. Each item is scored on a Likert scale from 1 “not affected” to 5 “most affected”, ranging the total score from 0 to 100. In the present study we used the total fatigue score. The Spanish version used in this study showed good internal consistency (Cronbach’s α = 0.83) [37]. In the sample of the present study, Cronbach’s alpha coefficient of the MFI was 0.948.

The Hospital Anxiety and Depression Scale (HADS) [38] includes 14-items (from 0 to 3 points) designed to assess symptoms of depression and anxiety in non-psychiatric hospital setting. Subscales are rated from 0 to 21, and the scores are categorized as follows: scores up to 7 are considered “normal”, scores from 8 to 10 indicate “possible clinical problems”, and scores above 11 suggest “clinical problems”. The Spanish adaptation presents good internal consistency for each subscale (Cronbach’s α ≥ 0.70) and test-retest reliability [39]. In the sample of the present study, Cronbach’s alpha coefficient of the HADS anxiety was 0.883, and for the HADS depression was 0.903.

The Fibromyalgia Impact Questionnaire (FIQ) [40] is a 10-item self-report that evaluates the current health status of patients with FM. Item 1 assesses functional capacity for daily living and is rated on a Likert scale from 0 to 3. Items 2 and 3 assess the number of days in the past week that the patients felt well or were unable to work, and are rated on a Likert scale from 0 to 7. Items 4 to 10 assess work difficulties, pain, fatigue, morning tiredness, stiffness, and mood state, each rated on a Likert scale from 1 to 10. The total score ranges from 0 to 100, and the cut-off points are as follows: scores ˂ 39 indicate mild impairment, scores ≥ 39 and ˂ 59 indicate moderate impairment, and scores ≥ 59 indicate severe impairment (44). In the CSS group, the patients answered the items considering their own CSS. The Spanish version of the FIQ has acceptable internal consistency (Cronbach’s α = 0.82), validity, and sensitivity to change [41, 42]. In the sample of the present study, Cronbach’s alpha coefficient of the FIQ was 0.897.

Pressure pain threshold was assessed using a digital algometer (Somedic AB, Farts, Sweden), applied by a trained nurse. The device features a 1 cm² rubber pointer that applies pressure perpendicularly to the selected anatomical site at a constant rate of 30 kPa/s. Measurements were performed bilaterally and alternately at the following anatomical sites: the tibial anterior muscle, the joint space between C5 and C6, and the intermetacarpal space between the second and third fingers, following the protocol of previous studies [37, 43]. Before the final measurements, two test trials were performed at anatomical regions not included in the study. Participants were instructed to verbally indicate “stop” when the sensation became painful. Subsequently, three measurements were taken at each site, with a 30-second interval between them, and the three values ​​were averaged to determine the pressure pain threshold.

Objective sleep quality was assessed using MotionWatch 8 actigraphs (CamNtech Ltd., Cambridge, UK) and MotionWare 1.1.15 software for data collection and analysis of sleep parameters, following the protocol described in a previous study [44]. The actigraph is a wrist-worn device that records daily activity levels. Participants wore the actigraphs continuously 24 hours a day (except during showering) for one week. The sleep parameters obtained (averaged values ​​over the week) were sleep efficiency, sleep latency, and sleep fragmentation index (sum of nocturnal mobility episodes and immobility episodes lasting ≤ 1 min). Actigraphy has demonstrated accuracy (86%) and sensitivity (90%) for assessing sleep parameters [45].

Data analysis

IBM SPSS Statistics (version 28.0.1.0) and the structural equation modeling toolbox of JASP software (version 0.17.2) were used. Statistical significance was considered at p < .05 with 95% confidence intervals. Basic descriptive statistics (mean, SD, and percentage) were presented for parameters related to traumatic experiences, PTSS, and the remaining clinical variables. To compare sociodemographic and clinical characteristics between groups (FM vs. CSS vs. healthy subjects), the χ² test was used for categorical variables and ANCOVA for quantitative variables (preceded by Levene’s test for equality of variances) controlling age. Tukey’s test (for equal variances) and Tamhane’s T2 test (for unequal variances) were used for post-hoc comparisons. Effect sizes were calculated using η² with the following guidelines [46]: small effect size (0.01 to 0.039), medium (0.06 to 0.11), and large (> 0.14).

To examine the influence of diagnostic group, age, and educational level on the presence of PTSD diagnosis, a binary logistic regression model was applied. The dependent variable was the presence of PTSD (0 = no, 1 = yes), and the independent variables were diagnostic group (FM, CSS and Healthy), age (continuous), and educational level (ordinal with three levels: low, medium, high). To assess the effects of the same variables on the PTSS intensity (continuous outcome), a General Linear Model (GLM) univariate was used. In this model, diagnostic group was included as a fixed factor, while age and educational level were treated as covariates. Homogeneity of variances was assessed using Levene’s test. Effect sizes were reported as η²_p for GLM and odds ratios for logistic regression.

In the FM group, trauma-exposed participants who developed PTSD (n = 40) were compared with those who did not develop PTSD (n = 35) on the various clinical symptoms using the Student t test for independent samples. Effect sizes were calculated using Cohen’s d according to the following guidelines [46]: small effect size (0.2 to 0.4), medium (between 0.5 and 0.7), and large (0.8 or greater). Besides, in the FM subgroup with PTSD, relationships between clinical measures were examined using Pearson’s correlation coefficient. Correlations were interpreted as follows: low (0.10 to 0.29), moderate (0.30 to 0.49), and high (0.50 or greater) [46]. In this subgroup, considering previous literature and the pattern of correlations identified, we analyze possible mediators between PTSS severity and central sensitization manifestations using a multiple mediation model. Direct effects, indirect effects (specific and total), and total effects were estimated [47]. The direct path refers to the effect of variable x on variable y. Specific indirect effects represent the path from x to y through a specific mediator. The total indirect effect of x on y is the sum of the specific indirect effects across all mediators. The total effect of x on y is the sum of the direct and indirect effects. When two or more variables are significant mediators of the relationship between x and y, indirect effects are significant in the mediation analysis. The minimum sample size required for regression analyses in the mediation model was n = 35 participants, considering an alpha level of 0.05, two mediators, a desired statistical power of 0.80, and an expected effect size of 0.30. The significance of effects was tested using delta method standard errors, bias-corrected percentile bootstrap confidence intervals, and the ML estimator. Estimates were based on 1.000 bootstrap samples, and 95% confidence intervals were considered.

Results

Sociodemographic and clinical characteristics

The sociodemographic variables are detailed in Table 1. The mean age of the FM group was significantly higher than that of the CSS group and the healthy group, F(2, 163) = 27.98, p < .001, η² = 0.256. Regarding marital status, the FM group showed a significantly greater presence of married people/living together compared to the CSS and the healthy group, χ²(8) = 53.05, p < .001. Also, we found a higher educational level in the CSS and healthy groups compared to the FM group, χ²(10) = 59.96, p < .001. Finally, as expected, the FM group showed a greater percentage of people retired/on sick leave, whereas the CSS and healthy groups were mainly active workers, χ²(10) = 85.26, p < .001). In the FM group, the average duration of pain was 7.65 years (SD = 7.36), and the main time since diagnosis was 7.38 years (SD = 7.55). These data are similar to those described in other studies with FM [2, 14].

Table 1.

Sociodemographic characteristics

Variable	FM			CSS			Healthy
	n (%)	M	SD	n (%)	M	SD	n (%)	M	SD	χ2 (gl,n)	F (2)	p	d	r	ƞ2
		N = 82			N = 43			N = 41
Age		53.2	8.54		38.16	15.07		40.71	14.07		27.98 (a,b)	< .001			0.256
Marital status										53.05 (8, 165)		< .001	1.377	0.567	0.321
Married	55 (67.9)			13 (30.2)			11 (26.8)
Separated	4 (4.9)			1 (2.3)			1 (2.4)
Divorced	13 (16)			2 (4.7)			4 (9.8)
Widow	3 (3.7)			0			1 (2.4)
Single	6 (7.4)			27 (62.8)			24 (58.5)
Educational level	68 (84)			14 (32,6)			16 (39)			59.96 (10, 165)		< .001	1.511	0.603	0.036
No schooling	1 (1.2)			0			0
Elementary	19 (23.2)			1 (2.3)			5 (12.5)
Secondary	22 (26.8)			2 (4.7)			3 (7.5)
High school	8 (9.8)			1 (2.3)			3 (7.5)
Professional	22 (26.8)			11 (25.6)			3 (7.5)
University studies	10 (12.2)			28 (65.1)			26 (65)
Employment status										85.26 (10, 164)		< .001	2.081	0.721	0.520
Student	2 (2.5)			14 (32.6)			3 (7.5)
Working	20 (24.7)			20 (46.5)			35 (87.5)
Retired	9 (11.1)			0			1 (2.5)
Unemployed	20 (24.7)			6 (14)			1 (2.5)
Sick leave	19 (23.5)			1 (2.3)			0
Disability leave	11 (13.6)			2 (4.7)			0

Note. *p < 0.05; **p < 0.01; ***p < 0.001; a = FM vs. CSS; b = FM vs. healthy; c = CSS vs. healthy; FM = Fibromyalgia group; CSS = Central Sensitization Syndrome group

Comparative analysis between groups

Trauma exposure and PTSD

Table 2 shows the characteristics of trauma, intensity of PTSS, and fulfillment of PTSD diagnostic criteria (DSM-5) in the three study groups. The FM group showed greater trauma severity, χ2 (6, N = 166) = 13.72, p = .033, η² = 0.094, and more presence of trauma on several occasions or repeatedly than the other groups, χ2 (6, N = 166) = 56.17, p < .001, η² = 0.390, while no significant differences were observed between groups in the age at which the event occurred. The most frequent types of trauma reported by the FM, CSS, and healthy control groups in total were vehicle accidents, other accidents, and natural disasters, followed by harassment situations, accidental or sudden death of a loved one, physical violence, and rape.

Table 2.

Frequency and characteristics of trauma and psychological impact in the study groups

Variable	FM			CSS			Healthy
	(n = 82)			(n = 43)			(n = 41)
	n (%)	M	SD	n (%)	M	SD	n (%)	M	SD	χ2 (gl)	F (2)	p	d	r	ƞ2
										N = 166
Age of trauma onset		27.46	16.61		24.11	14.71		29.73	13.54		1.157	0.318
Severity of trauma										13.72 (6)		0.033	0.646	0.307	0.094
Mild	2 (2.8)			3 (7.9)			2 (5.6)
Moderate	5 (7)			4 (10.5)			6 (16.7)
Severe	26 (36.6)			23 (60.5)			16 (44.4)
Extreme	38 (53.5)			8 (21.1)			12 (33.3)
Frequency of trauma										56.17 (6)		< .001	1.599	0.624	0.390
Never	3 (4.2)			20 (51.3)			19 (55.9)
Once	27 (38)			9 (23.1)			11 (32.6)
Several times	16 (22.5)			9 (23.1)			4 (11.8)
Repeatedly	25 (35.2)			1 (2.6)			0
Criterion A	62 (96,9)			32 (94.1)			62 (87.9)			3.07 (2)		0.215	0.310	0.153	0.023
Criterion B	63 (84)			33 (82.5)			31 (75.6)			1.27 (2)		0.529	0.181	0.09	0.008
Criterion C	54 (73)			22 (55)			20 (48)			7.65 (2)		0.022	0.455	0.222	0.049
Criterion D	59 (79,7)			21 (52.5)			18 (43.9)			17.23 (2)		< .001	0.707	0.333	0.111
Criterion E	62 (83,8)			26 (65)			19 (46.3)			17.71 (2)		< .001	0.718	0.338	0.114
Criterion F	64 (100)			32 (100)			29 (93.5)			6.29 (2)		0.043	0.456	0.222	0.049
Criterion G	50 (67,6)			18 (46.2)			7 (17.1)			27.06 (2)		< .001	0.923	0.419	0.175
PTSD diagnosis	40 (53,3)			12 (30)			0			34.2 (2)		< .001	1.059	0.468	0.219
Depersonalization	18 (24,3)			7 (17.5)			0			11.6 (2)		0.003	0.569	0.273	0.074
Derealization	32 (43,2)			9 (23.1)			0			25.59 (2)		< .001	0.892	0.407	0.166
Delayed onset of symptoms	9 (14,3)			3 (9.4)			1 (3.2)			2.78 (2)		0.248	0.300	0.148	0.022
Type of trauma
Natural disasters	60 (76.9)			33 (84.6)			32 (94.1)			5.03 (2)		0.081	0.371	0.182	0.033
Vehicle accidents	65 (83.3)			38 (92.7)			36 (94.7)			4.21 (2)		0.122	0.332	0.163	0.026
Other accidents	58 (74.4)			37 (90.2)			32 (94.1)			8.63 (2)		0.013	0.489	0.237	0.056
War	34 (43.6)			26 (68.4)			26 (78.8)			14.16 (2)		< .001	0.648	0.308	0.095
Rape	40 (51.3)			28 (68.3)			24 (77.4)			7.54 (2)		0.023	0.460	0.224	0.050
Harassment	54 (69.2)			35 (85.4)			31 (86.1)			6.02 (2)		0.049	0.402	0.197	0.038
Physical violence	47 (60.3)			31 (75.6)			27 (84.4)			7.21 (2)		0.027	0.447	0.218	0.047
Acts of terrorism	26 (33.3)			22 (56.4)			20 (66.7)			11.88 (2)		0.003	0.593	0.284	0.080
Imprisonment	35 (44.9)			25 (61)			19 (65.5)			4.93 (2)		0.085	0.371	0.182	0.033
Accidental death	53 (68.8)			32 (80)			32 (86.5)			4.74 (2)		0.093	0.356	0.175	0.03
Other	46 (63.9)			24 (70.6)			23 (74.2)			1.2 (2)		0.547	0.188	0.093	0.008
Total symptom intensity		29.6	20.54		20.73	19.02		13.38	14.56		11.82 (a,b)	< .001			0.119
Intrusive symptoms		7.92	5.98		5.43	5.02		4.63	4.79		5.59 (b)	0.005			0.069
Avoidance		2.93	2.74		2.59	3.02		1.62	2.23		3.80 (b)	0.026			0.04
Cognitive and mood disturbances		9.47	7.66		6.35	7.89		3.6	4.77		12.39 (b)	< .001			0.109
Hyperarousal and reactivity		9.47	6.48		6.43	5.97		3.58	4.66		13.15 (a,b)	< .001			0.149
Functional impact		3.16	2.33		1.97	2		0.61	0.97		35.77 (a,b,c)	< .001			0.228

Note. a = FM vs. CSS; b = FM vs. healthy; c = CSS vs. healthy; Criterion A = Trauma exposure; Criterion B = Intrusive symptoms; Criterion C = Avoidance; Criterion D = Cognitive and mood disturbances; Criterion E = Hyperarousal and reactivity; Criterion F = Symptom duration; Criterion G = Functional impact; PTSD = Post-traumatic stress disorder; FM = Fibromyalgia group; CSS = Central Sensitization Syndrome group

Regarding PTSD diagnose, the FM group shows a higher presence of cases diagnosed with PTSD compared to the CSS and healthy groups χ2 (2, N = 166) = 34.2, p < .001, η² = 0.219; as well as a significantly higher prevalence of symptoms of depersonalization χ2 (2, N = 166) = 11.61, p = .003, η² = 0.074, and derealization, χ2 (2, N = 166) = 25.59, p < .001, η² = 0.166. No group differences were found for the delayed expression criterion. Also, PTSS intensity was significantly higher in the FM group compared to the CSS and healthy groups, F(2, 150) = 11.82, p < .001, η² = 0.119, and no significant differences were found between the CSS and healthy groups in this regard. The FM group had significantly higher hyperarousal intensity, F(2, 150) = 13.15, p < .001, η² = 0.149, than the other groups. The functional impact of PTSD was significantly highest in the FM group, followed by the CSS and the healthy group, F(2, 151) = 35.77, p < .001, η² = 0.228. Furthermore, the FM group showed greater intensity of intrusive symptoms, F(2, 150) = 5.59, p = .005, η² = 0.069, avoidance, F(2, 147) = 3.80, p = .026, η² = 0.040; and cognitive and mood disturbances, F(2, 149) = 12.39, p < .001, η² = 0.109, than the healthy group, although there were no significant differences in these symptoms between the FM and CSS groups, nor between the CSS and healthy controls. According to the cut-off points of the EGEP-5 questionnaire, PTSS intensity was moderate in the FM and CSS groups, and low in the healthy group.

To examine the influence of diagnostic group, age, and educational level on the presence of PTSD diagnosis, a binary logistic regression model was applied, and to assess the effects of the same variables on the PTSS intensity, a GLM univariate was used. Regarding PTSD diagnosis, the logistic regression model was statistically significant (χ² = 5.02, p < .001). The Hosmer-Lemeshow goodness-of-fit test showed a good model fit (p = .955). The diagnostic group significantly predicted PTSD diagnosis (χ² = 7.57, p = .023), as did educational level (χ² = 4.70, p = .096), with significant contrasts for low and medium levels. Age was not a significant predictor (p = .244). Compared to the control group, participants with FM showed significantly higher odds of having a PTSD diagnosis (OR = 5.02, p = .006), while the SSC group did not differ significantly (p = .997).

The GLM univariate revealed a significant effect of diagnostic group on PTSS intensity (F(2,147) = 11.50, p < .001, η²_P = 0.135). Neither age (p = .138) nor educational level (p = .406) had significant effects. Post hoc pairwise comparisons showed that participants with FM reported significantly higher PTSD symptom intensity (M = 31.67, SD = 2.46) compared to both the SSC group (M = 18.43, SD = 3.22; p = .009) and the control group (M = 11.87, SD = 3.11; p < .001). No significant differences were found between the SSC and control groups (p = .374).

Other clinical symptoms

After adjusting for age, significant group differences were found in sensory-affective pain, and pain intensity (self-report), as measured by the McGill Pain Questionnaire. The FM group reported the highest scores, followed by the CSS group and the healthy group, F(2, 160) = 16.12, p < .001, η²_P = 0.168, and F(2, 160) = 28.02, p < .001, η²_P = 0.259, respectively, (please, see Table 3). Age and the group × age interaction were not significant in either model (ps > 0.20). Similarly, pressure pain thresholds, assessed via algometry, were significantly lower in the FM group compared to both the SSC and healthy groups, F(2, 160) = 8.23, p < .001, η²_P = 0.093. No significant effects were observed for age or the group × age interaction (ps > 0.09).

Table 3.

Comparison between the groups in clinical characteristics

Variable	FM		CSS		Healthy
	(n = 82)		(n = 43)		(n = 41)
	M	SD	M	SD	M	SD	F (2)	p	ƞ2p
Cognitive-affective pain (SF-MPQ)	30.19	1.27	13.26	1.54	1.92	1.49	16.12 (a,b,c)	< .001	0.168
Pain intensity (VAS, SF-MPQ)	8.39	0.254	5.51	0.308	1.63	0.298	28.02 (a,b,c)	< .001	0.259
Central sensitization (CSI)	75.00	1.61	49.05	1.96	28.24	1.89	27.34 (a,b,c)	< .001	0.255
Fatigue (MFI)	79.33	1.97	59.41	2.39	42.35	2.32	10.13 (a,b,c)	< .001	0.112
Disability (FIQ)*1	71.42	15.74	43.94	15.75			9.24	0.003	0.071
Sleep quality (PSQI)	15.98	0.560	11.29	0.679	6.11	0.658	7.01 (a,b,c)	< .001	0.081
Depression (HADS)	11.78	0.625	6.24	0.758	3.47	0.735	4.55 (a,b,c)	0.012	0.054
Anxiety (HADS)	13.38	0.622	9.07	0.754	6.33	0.731	5.18 (a,b,c)	0.007	0.061
Pain catastrophizing (PCS)	31.65	1.722	20.76	2.088	10.10	2.024	4.10 (a,b,c)	0.018	0.049
Pain acceptance (CPAQ)	43.47	2.900	59.74	3.518	71.46	3.409	5.27 (a,b,c)	0.006	0.063
Pain threshold (algometer)	1.37	0.219	3.50	0.266	4.24	0.258	8.23 (a,b)	< .001	0.093
Sleep efficiency (actigraphy)	85.54	0.593	84.32	0.767	86.20	0.708	0.037	0.963	0.001
Sleep latency (actigraphy)	993.60	108.74	761.00	140.68	811.57	129.89	0.450	0.638	0.006
Seep fragmentation (actigraphy)	24.43	1.07	27.77	1.39	23.03	1.28	4.24 (b)	0.016	0.053

Note. FM = Fibromyalgia group; CSS = Central Sensitization Syndrome group; Healthy = Healthy group; a = FM vs. CSS; b = FM vs. healthy; c = CSS vs. Healthy; SF-MPQ = Short-Form McGill Pain Questionnaire; VAS = Visual Analogue Scale; CSI = Central Sensitization Inventory; MFI = Multidimensional Fatigue Inventory; PSQI = Pittsburgh Sleep Quality Index; HADS = Hospital Anxiety and Depression Scale; PCS = Pain Catastrophizing Scale; CPAQ = Chronic Pain Acceptance Questionnaire; *1 = The FIQ was administered only in the FM and CSS groups

Significant differences between the groups were also observed in the CSI, F(2, 160) = 27.34, p < .001, η²_P = 0.255. While the covariate age was not significant, the interaction between group and age was, F(2, 160) = 3.27, p = .040, suggesting that age may moderate the expression of central sensitization symptoms across groups. According to established cut-off points, CSI were extreme in the FM group, moderate in the CSS group, and subclinical in the healthy control group. Group differences were also found in fatigue F(2, 160) = 10.13, p = .001, η²_P = 0.112, with the FM group reporting the highest fatigue levels. No effects of age or the group × age interaction was found (ps > 0.11). Functional disability was significantly greater in the FM group than in the CSS group, F(1, 121) = 9.24, p = .003, η²_P = 0.071, with the FM group displaying scores indicative of severe impairment and the CSS group falling in the moderate range. No effects of age or the group × age interaction was found (ps > 0.11).

Regarding sleep quality, significant differences were observed between groups in the self-report measure, F(2, 160) = 7.01, p = .001, η²_P = 0.081, with the FM group reporting the poorest sleep quality. Neither age nor the group × age interaction were significant (ps > 0.14). Objective measures via actigraphy showed no significant group differences in sleep efficiency or latency Fs(2, 150) < 1, ps > 0.60. However, a significant effect of group was observed in the fragmentation index, F(2, 150) = 4.24, p = .016, η²_P = 0.053, with greater fragmentation in the clinical groups. Notably, both age (F(1, 150) = 9.57, p = .002) and the group × age interaction (F(2, 150) = 5.34, p = .006) were significant, suggesting a differential pattern of sleep disruption across groups depending on age.

Finally, in the domain of cognitive-affective symptoms, the FM group reported the highest levels of anxiety, F(2, 160) = 5.18, p = .007, η²_P = 0.061, and depression, F(2, 160) = 4.55, p = .012, η²_P = 0.054. In both models, neither age nor the group × age interaction were significant (ps > 0.36). The FM group scored in the clinical range, CSS in the borderline range, and controls in the normal range. Pain catastrophizing was significantly elevated in the FM group compared to the other groups, F(2, 160) = 4.10, p = .018, η²_P = 0.049. Pain acceptance was significantly lower in the FM group, F(2, 160) = 5.27, p = .006, η²_P = 0.063, with a linear pattern across groups: FM < CSS < healthy. Neither age nor the group × age interaction effects were significant for these variables (ps > 0.08).

Comparisons of fibromyalgia subgroups with and without PTSD diagnosis

The FM subgroup with PTSD had significantly higher scores on central sensitization symptoms, t(73) = -1.72, p = .045, d = -0.398; fatigue, t(73) = -2.13, p = .018, d = -0.492; anxiety, t(73) = -2.44, p = .009, d = -0.560; pain catastrophizing, t(73) = -2.16, p = .017, d = -0.500 and low pain acceptance, t(73) = 2.28, p = .013, d = 0.528, compared to the FM subgroup without PTSD (please, see Table 4). On the contrary, no statistically significant differences were found between the FM group that developed PTSD and those that did not in pain severity, pain pressure thresholds, disability, depression, subjective sleep quality, and actigraphy parameters, although scores tended to indicate greater impairment in the FM group with PTSD.

Table 4.

Clinical characteristics of FM group with PTSD vs. FM group without PTSD

Variable	FM without PTSD		FM with PTSD
	(n = 35)		(n = 40)
	M	SD	M	SD	t(gl)	p	d
Cognitive-affective pain (SF-MPQ)	27.22	12.12	28.75	9.33	-0.602 (63.5)	0.274	-0.142
Pain intensity (VAS, SF-MPQ)	8.17	1.31	8.07	1.43	0.301 (73)	0.382	0.070
Central sensitization (CSI)	71.26	10.69	75.18	9.05	-1.718 (73)	0.045	-0.398
Fatigue (MFI)	77.11	12.70	82.85	10.65	-2.126 (73)	0.018	-0.492
Disability (FIQ)	68.6	16.32	73.26	15.55	-1.267 (73)	0.105	-0.293
Sleep quality (PSQI)	15	4.08	16.37	3.39	-1.591 (73)	0.058	-0.368
Depression (HADS)	11.17	4.87	12.68	4.58	-1.377 (73)	0.086	-0.319
Anxiety (HADS)	11.71	3.96	14.10	4.44	-2.440 (73)	0.009	-0.565
Pain catastrophizing (PCS)	27.71	11.75	33.45	11.22	-2.160 (73)	0.017	-0.500
Pain acceptance (CPAQ)	50.94	20.44	40.58	18.93	2.279 (73)	0.013	0.528
Pain threshold (algometer)	1.42	0.98	1.31	0.85	0.514 (73)	0.304	0.119
Sleep efficiency (actigraphy)	85.39	5.45	85.67	3.35	-0.260 (53)	0.398	-0.063
Sleep latency (actigraphy)	1038.35	1110.35	912.48	660.91	0.578 (52)	0.283	0.140
Sleep fragmentation (actigraphy)	27.74	10.79	26.46	7.42	0.595 (71)	0.277	0.140

Note. FM = Fibromyalgia group; PTSD = Posttraumatic stress disorder; SF-MPQ = Short-Form McGill Pain Questionnaire; VAS = pain Visual Analogue Scale; CSI = Central Sensitization Inventory; MFI = Multidimensional Fatigue Inventory; FIQ = Fibromyalgia Impact Questionnaire; PSQI = Pittsburgh Sleep Quality Index; HAD = Hospital Anxiety and Depression Scale; PCS = Pain Catastrophizing Scale; CPAQ = Chronic Pain Acceptance Questionnaire

Relationships between traumatic experiences, PTSS and clinical variables

As Table 5 shows, in the subgroup of FM with PTSD, the number of traumatic experiences correlated significantly with central sensitization (r = .37, p = .019), disability (r = .35, p = .028) and poor sleep efficiency (actigraphy) (r = − .32, p = .045), but did not correlate with any of the PTSS subscales or with the rest of the physical and psychological variables considered (all’s p > .131). The intensity of PTSS significantly correlated with sensory-affective pain (r = .32, p = .047), CS (r = .51, p < .001), fatigue (r = .36, p = .023), anxiety (r = .32, p = .045), and pain acceptance (r = − .33, p = .035), with the cognitive and mood disturbances subscale (PTSD) showing a similar pattern of associations. Other correlations can be seen in Table 5, highlighting the number and magnitude of the correlations of the CSI with all measures.

Table 5.

Correlations among number of Traumas, PTSS severity and Self-Reports in FM group with PTSD

Variable	Number of traumas	PTSS Intensity Total	Intrusive symptoms	Avoidance behaviours	Cognitive and mood disturbances	Hyperarousal and reactivity (EGEP-5)
	(EGEP-5)	(EGEP-5)	(EGEP-5)	(EGEP-5)	(EGEP-5)
Sensory-affective pain (SF-MPQ)	0.225	0.315*	0.209	0.162	.350*	0.280
Pain intensity (VAS, SF-MPQ)	0.146	0.264	0.302	0.132	0.245	0.279
Central sensitization (CSI)	0.369*	0.511***	0.316*	0.341*	0.562***	0.433**
Fatigue (MFI)	-0.108	0.358*	0.380*	0.381*	0.277	0.225
Disability (FIQ)	0.347*	0.282	0.230	0.071	.394*	0.240
Sleep quality (PSQI)	0.123	0.243	0.222	0.224	0.202	0.192
Depression (HADS)	0.094	0.243	0.068	-0.036	0.360*	0.239
Anxiety (HADS)	0.202	0.319*	0.157	0.012	0.296	0.404**
Pain catastrophizing (PCS)	0.002	0.302	0.184	0.207	0.267	0.224
Pain acceptance (CPAQ)	0.243	− 0.334*	-0.204	-0.161	− 0.452**	-0.311
Pain threshold (algometer)	-0.147	-0.105	-0.152	-0.198	-0.091	0.038
Sleep efficiency (actigraphy)	−0.323*	0.022	0.135	-0.073	0.053	0.098
Sleep latency (actigraphy)	0.082	0.107	0.057	0.096	0.099	0.041
Sleep fragmentation (actigraphy)	0.121	-0.065	-0.075	-0.022	-0.108	-0.126

Note. EGEP-5 = Global Assessment of Post-traumatic Stress Questionnaire; MPQ = Short Form-McGill Pain Questionnaire; VAS = Visual Analogue Scale; CSI = Central Sensitization Inventory; MFI = Multidimensional Fatigue Inventory; FIQ = Fibromyalgia Impact Questionnaire; PSQI = Pittsburgh Sleep Quality Index; HAD = Hospital Anxiety and Depression Scale; PCS = Pain Catastrophizing Scale; CPAQ = Chronic Pain Acceptance Questionnaire. *p < .05. **p < .01, ***p < .001

Mediators in the association between PTSS and central sensitization

Based on previous literature and the correlations pattern observed, the role of anxiety and pain acceptance as mediators of the effect of PTSS on central sensitization was examined (see Table 6; Fig. 1). The direct effect of PTSS on central sensitization was significant (z = 2.47, p = .014). The total indirect effects of PTSS on central sensitization via the two mediators were significant (z = 2.31, p = .021), and the specific indirect effects via anxiety and pain acceptance were close to significance (z = 1.69, p = .090 and z = 1.66, p = .097, respectively). The total effect (direct + indirect) of PTSS on central sensitization was significant (z = 3.77, p < .001), explaining this model the 21% of the variance.

Table 6.

Effect of PTSS on CS through anxiety and pain acceptance in FM group with PTSD

	Estimate	SE	z-value	p	95% CI
					Lower	Upper
Direct effects
PTSS → CS	0.020	0.008	2.470	0.014	0.006	0.033
Indirect effects
PTSS → Anxiety → CS	0.007	0.004	1.694	0.090	9.763×10− 4	0.017
PTSS → Pain acceptance → CS	0.007	0.004	1.659	0.097	6.062×10− 4	0.018
Total	0.014	0.006	2.313	0.021	0.003	0.028
Total effects
PTSS → CS	0.034	0.009	3.765	< .001	0.020	0.049

Note. Delta method standard errors, bias-corrected percentile bootstrap confidence intervals, ML estimator

Fig. 1.

Mediation model linking PTSS and CS in FM group with PTSD. Note. *p < .05. **p < .01, ***p < .001

Discussion

The aim of this study was to explore the link between traumatic experiences and central sensitization, comparing participants with FM, CSS and healthy people in PTSS/PTSD and other clinical characteristics, and attempting to identify mediators between PTSS and central sensitization intensity. We found that the FM group showed a higher frequency and severity of trauma, with more presence of trauma on several occasions or repeatedly, than the CSS and healthy groups, while the groups did not differ in the age at which the event occurred. Previous retrospective studies have found a higher self-reported frequency and severity of trauma in FM compared to healthy and other chronic pain patients as rheumatoid arthritis [9, 10, 13, 18, 20, 48], and in CSS versus healthy and other chronic pain conditions [9, 10, 15, 49], to which our data add that within CSS the FM group would be the most affected. Furthermore, although several studies have highlighted the association between early childhood trauma and FM or CSS [9, 14], we found that the traumas reported by the three study groups can occur at any age with a predominance of adolescence and adulthood. This result seems consistent with the relevance of lifelong victimization, chronic stress being more relevant than acute stress [10, 13, 20, 50].

The most frequently reported types of traumas across the three study groups were accidents, mainly vehicle-related and natural disasters, followed by harassment, accident or sudden death of a family member or close person, physical violence, and rape. Previous research has highlighted the association between both physical traumas, particularly vehicle-related accidents, and emotional trauma with FM [11, 14, 18, 20, 48, 49] or CSS [10, 15, 50], with reported prevalence rates ranging from 20 to 70% depending on the study. Overall, these findings are consistent with what was observed in our study.

This greater frequency and severity of trauma in the FM group is also accompanied by a higher incidence of PTSD diagnoses compared to the CSS and healthy groups (53.3%, 30% and 0%, respectively), as well as greater PTSS intensity and higher prevalence of depersonalization and derealization, which is consistent with our first hypothesis. Specifically, the FM group showed greater intensity of intrusions, avoidance, and cognitive and mood disturbances, than the healthy group, although there were no differences in these symptoms between the FM and CSS groups. Besides, the FM group showed higher hyperarousal intensity and functional impairment due to PTSD than the other groups. These findings are consistent with the studies that have analyzed the presence of PTSD in groups with FM [14], or CSS [15, 16], where PTSD rates range between 35 and 50%. These rates are much higher than those reported in the general population (e.g. around 4%) [49]. The exact prevalence rate of PTSD varies depending on the assessment instrument used in each study and the characteristics of the participants. The high PTSD rate observed in our FM group may be related to the fact that our sample was collected from clinical settings.

The FM group also showed greater severity in the remaining clinical symptoms evaluated followed by the CSS group and healthy individuals. Thus, the group with FM showed higher pain (self-report of sensory-affective aspects and pain intensity and lower pain thresholds in the algometer), central sensitization, fatigue, disability, anxiety, depression, pain catastrophizing and low pain acceptance than the other groups, which is consistent with the high level of deterioration and comorbidity that characterizes FM [2, 51]. Among all the clinical symptoms evaluated, only the level of central sensitization exhibited an age-related effect, specifically within the FM group. Higher age was associated with increased level of central sensitization, a finding that aligns with the notion that, over time, sensitization phenomena within neural networks involved in pain perception may become more chronic and severe [6, 7]. Similarly, FM patients reported worse subjective sleep quality than CSS and healthy participants. In the actigraphy, the FM group showed greater sleep fragmentation compared to the healthy group, although no significant differences were observed between the FM and CSS groups. No group differences were found in actigraphy-based measures of sleep latency and sleep efficiency. It is estimated that between 50% and 90% of individuals with CSS experience sleep disorders, especially those with FM where the rates reach 94.7–99% of cases [37, 52], which aligns with our findings. Sleep fragmentation is the most relevant indicator of sleep disturbances detected by actigraphy and has been correlated with a state of hyperactivation prior to sleep, which makes sleep more superficial [52, 53]. In the FM group, sleep fragmentation also tends to increase with age, consistent with the age-related developmental changes that occur in sleep, which are characterized by a decrease in the proportion of deep slow-wave sleep and an increase in light sleep phases and the number of awakenings [52, 53]. On the other hand, despite its usefulness, actigraphy results do not always coincide with self-reported sleep quality. It has been observed that individuals tend to underestimate the duration and subjective quality of sleep compared to actigraphy data, especially when their sleep quality is poor [53]. This discrepancy may help explain why, in our findings, self-reported sleep quality is markedly affected, whereas actigraphy only reveals differences in sleep fragmentation. This pattern has also been reported in another study that assesses sleep quality in FM using actigraphy [44].

The FM subgroup who develops PTSD, compared to FM subgroup without PTSD, showed a higher level of central sensitization, fatigue, anxiety, pain catastrophizing and pain acceptance. No differences were observed among these groups in pain severity, pain pressure thresholds, disability, depression, subjective sleep quality, and actigraphy parameters, although scores tend to reflect greater deterioration in the FM subgroup with PTSD. The studies exploring the presence of PTSS/PTSD in FM, clearly show a mutual exacerbation of symptoms, generating greater pain and somatization, emotional distress, dysfunctional emotional coping, fatigue and disability [14, 15, 16, 20, 48, 54], and poor subjective sleep quality [15, 20]. Although comparative studies including pain catastrophization and pain acceptance, or actigraphy are lacking. Thus, our findings partially agree with existing literature and our second hypothesis. The lack of differences in pain, disability, depression, and sleep quality could be due to the high scores in these measures in the FM group, which may have generated a ceiling effect, although further research is needed in this area.

In addition, the pattern of correlations in the subgroup of FM with PTSD reflects somewhat similar results. Thus, the intensity of PTSS was associated with the central sensitization index, sensory-affective aspects of pain, fatigue, anxiety, and pain acceptance, with the cognitive and mood disturbances subscale (PTSD) showing similar associations. Instead, the number of traumatic experiences correlated with central sensitization, disability, and poor sleep efficiency (actigraphy). The CSI correlates with both the frequency of trauma and the intensity of PTSS and each of its specific symptoms. It is well established that experiencing trauma can negatively affect both physical and psychological functioning, as reflected in our findings through the correlations between trauma frequency, central sensitization, sleep quality, and disability, in line with previous studies [10, 50]. However, studies examining the presence of PTSS/PTSD diagnostic criteria in FM [14], or other chronic pain conditions [15, 16] show that the impact of trauma is up to four times greater than the trauma itself. This finding is consistent with the stronger pattern of correlations observed in our results between the PTSS intensity and clinical variables.

Based on previous literature and the observed pattern of correlation, we examined the role of anxiety and pain acceptance as mediators of the effect of PTSS on central sensitization. The results indicated that both anxiety and pain acceptance were significant mediators in this relationship, accounting for 21% of the variance, in line with our third hypotheses. It has been assumed that the presence and maintenance of PTSD links experiencing trauma and developing pain or central sensitization [16, 55]. However, the underlying mechanism connecting having PTSS/PTSD with pain remains unknown, and only few studies have explored potential mediators or modulators in this relationship. In our findings, anxiety emerged as a significant mediator between PTSS and central sensitization, consistent with previous findings showing that anxiety —but not sleep quality or coping styles— achieves a mediating role in the relationship between PTSS and daily functioning [20]. In contrast, although one study found that the relationship between PTSS after sexual abuse and pain was mediated by depression, the other existing studies do not support a mediating role for depression in the relationships between PTSS and FM [19, 54]. It is estimated that at least one-quarter of FM patients have depression or anxiety disorder [56], and symptoms prevalence is substantially higher, for example: up to 87.5% show manifestations of anxiety, and 76.1% depressed mood [14]. Emotional distress may precede and/or follow pain: although depression tends to follow pain, and anxiety tends to precede pain onset [22]. Furthermore, this mediating role of anxiety seems to be consistent with the current etiological models of FM, which suggest dysregulation of the hypothalamic-pituitary-adrenal axis activity due to chronic stress. This dysregulation may impair the endocrine, sympathetic, and immune systems, resulting in a state of central sensitization [6, 13, 57]. Also, the recently proposed integrative Fibromyalgia Imbalance of Threat and Soothing Systems model [8] hypothesized that an imbalance in emotion regulation, reflected by an overactive “threat” system and underactive “soothing” system, might keep the “salience network” (or midcingulo-insular network) in a continuous state of alert. This hyperactivation, closely related to anxiety, contributes to FM or other CSS.

On the other hand, although no previous studies have analyzed the mediating role of pain-related attitudes in the relationship between PTSS/PTSD and FM or central sensitization, the mediating role of pain acceptance found in our study may contribute to perpetuating a state of anxiety and hyperactivation that could lead to central sensitization. Past traumatic experiences may give rise to dysfunctional cognitive-affective and interpersonal characteristics, such as early maladaptive schemas, which have been linked to dysfunctional pain-related attitudes [21], and to emotional dysregulation that increase anxiety and negative affect [58, 59]. Similar to pain catastrophizing, low pain acceptance correlates with increased emotional distress [60]; whereas acceptance has been identified as a key variable related to resilience [61]. Difficulties in emotional regulation and the tendency to reject or resist pain and distress may perpetuate a physiological state of hyperactivation, contributing to the emergence of central sensitization. This hypothesis is also consistent with the previously mentioned Imbalance of Threat and Soothing Systems Model, which suggests that, following adversity, the “salience network” remains in a continuous state of alert, thereby promoting the development of CSS [8, 57].

The present study has several strengths, including its multigroup design, the combined use of subjective and objective measures, particularly the continuous monitoring of sleep quality using actigraphy over a full week, and the application of rigorous diagnostic criteria for both FM and PTSD. However, the study also present several limitations that may affect the validity and generalizability of the results. First, although the effect of age was controlled, since the FM sample was slightly older than the CSS sample, the menopausal status of the patients was not assessed. While age control may partially account for this factor, further studies are needed to specifically evaluate the potential influence of menopause on clinical outcomes. In addition, the FM and CSS group differed in other sociodemographic variables that may have influenced the results. Nevertheless, the sociodemographic characteristics of the FM group are consistent with that typically observed in this population [2, 14], providing ecological validity and clinical relevance to the findings. Second, it should be noted that most individuals with FM also present comorbidity with other CSS, and that the CSS group included multiple diagnoses, making it heterogeneous by nature. Third, the study did not include psychophysiological pain measures, which limit the possibility to explore the functioning of ascending and descending pain modulation pathways and thus restrict a more comprehensive understanding of the nociceptive system. Finally, the cross-sectional design limits the possibility of establish causal relationships, as it does not allow for evaluate changes over time or determining the directionality of associations between variables. Future research should consider exploring additional subgroup comparisons across different CSS to further validate and refine the transdiagnostic findings reported in this study. Moreover, longitudinal studies with larger samples and a greater number of psychophysiological measures, such as cortisol levels or polysomnography, are needed to gain deeper understanding of the causality and underlying mechanisms in the relationship between PTSD and FM/CSS.

Conclusion

The present study indicated that FM patients showed a higher frequency and severity of trauma, and a higher incidence of PTSD diagnoses and PTSS intensity than the CSS and healthy participants. The group of FM showed greater severity in the remaining clinical symptoms followed by the CSS group and healthy individuals. Besides, the subgroup of people with FM who develop PTSD showed a more severe clinical profile compared to FM subgroup without PTSD, highlighting the correlations between the intensity of PTSS and several clinical symptoms. The results of the multiple mediation model provide additional evidence of the impact of PTSS on FM. We found that the relationship between PTSS intensity and central sensitization severity is mediated by anxiety and pain acceptance. These results may have potential implications for clinical practice, highlighting the importance of assessing PTSS/PTSD in FM and CSS patients. Our findings suggest that people with FM/CSS may require a tailored treatment approach, which emphasizes addressing emotions such as anxiety and increasing acceptance. In fact, the multimodal approach for the treatment of PTSD and chronic pain—which combines standard cognitive-behavioral therapy with acceptance and commitment therapy and mindfulness, aimed at improving acceptance of physical symptoms, along with emotional processing therapies for PTSD— has shown promising preliminary results [58]. Resolving trauma may have a relevant impact on FM/CSS and contribute substantially to improving the quality of life in this complex patient population.

Abbreviations

CPAQ

Chronic Pain Acceptance Questionnaire

CSI

Central Sensitization Inventory

CSS

Central sensitization syndromes

DSM-5

Diagnostic and Statistical Manual of Mental Disorders (5th ed.)

EGEP-5

Global Assessment of Post-Traumatic Stress Questionnaire

FIQ

Fibromyalgia Impact Questionnaire

FM

Fibromyalgia

HADS

Hospital Anxiety and Depression Scale

MFI

Multidimensional Fatigue Inventory

MPQ-SF

McGill Pain Questionnaire-Short form

PCS

Pain Catastrophizing Scale

PSQI

Pittsburgh Sleep Quality Index

PTSD

Post-traumatic stress disorder

PTSS

Post-traumatic stress symptoms

VAS

Visual analogue scale

Author contributions

C.M. conducted searches, analysed data, and drafted the manuscript. E.M. drafted the protocol, conducted searches, analysed data, and drafted and reviewed the manuscript. A.I.S. drafted the protocol, provided supervision, guided the research question, assisted with clinical interpretation of results, and reviewed manuscript drafts. R.C. Conducted patient selection and referral and assisted with clinical interpretation of results. M.P.M. drafted the protocol, analysed data, provided supervision, guided the research question, assisted with clinical interpretation of results, and reviewed manuscript drafts.

Funding

This study is part of a broader research project supported by the Spanish Ministry of Science and Innovation and Spanish State Research Agency (project PID2019-109612GB-I00, funded by MCIN/AEI/10.13039/501100011033), the María de Maeztu Unit of Excellence Program funded by the Spanish Ministry of Science, Innovation and Universities through the State Research Agency (CEX2023-001312-M /AEI/10.13039/501100011033) and the Unit of Excellence funded by the University of Granada (UCE-PP2023-11 /UGR).

Data availability

The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

This study was approved by the Provincial Ethics Committee of Granada (session held on November 25, 2019, minute 11/19) of the Council of Health and Family (Government of Andalusia, Spain). All participants were informed about the study’s objectives and signed informed consent forms. This study was conducted in accordance with the ethical principles outlined in the Declaration of Helsinki.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.