Altered somatosensory temporal discrimination in chronic non-specific low back pain patients with central sensitization-related features: a cross-sectional study

Abstract

Background

Central sensitization (CS) involves impaired inhibitory control and maladaptive neuroplasticity, resulting in altered sensory processing beyond nociceptive pathways. This study examined whether CS-related features are associated with impaired somatosensory temporal discrimination threshold (STDT) and its relationship with pain catastrophizing and disability in patients with chronic non-specific low back pain (cNSLBP).

Methods

Patients with cNSLBP were classified with or without CS-related features and compared with age- and sex-matched healthy controls (n = 37 per group). STDT was assessed at both the lumbar region and the hand. Pain catastrophizing, emotional status, and disability were evaluated using the Pain Catastrophizing Scale, the Hospital Anxiety and Depression Scale, and the Roland–Morris Disability Questionnaire. Between-group comparisons and correlation analyses were performed with adjustment for potential confounders.

Results

Patients with CS-related features demonstrated significantly prolonged STDT at both the lumbar region (F[2,101] = 33.80; P < 0.001; partial η² = 0.401) and the hand (F[2,101] = 50.20; P < 0.001; partial η² = 0.499) compared with patients without CS-related features and healthy controls. Lumbar STDT was associated with both pain catastrophizing (r = 0.529; P = 0.001) and disability (r = 0.343; P = 0.038), whereas hand STDT was associated exclusively with pain catastrophizing (r = 0.330; P = 0.046).

Conclusions

CS-related features in cNSLBP are associated with impaired somatosensory temporal discrimination at both the lumbar region and the hand, suggesting that non-nociceptive temporal processing deficits may reflect a generalized alteration in central sensory processing rather than a localized phenomenon.

INTRODUCTION

Chronic low back pain (CLBP) is one of the most common musculoskeletal disorders worldwide, which affects quality of life, leads to loss of productivity, and imposes a significant economic burden on healthcare systems [1]. In chronic non-specific low back pain (cNSLBP), the major subgroup of CLBP, patients’ symptoms cannot be adequately explained by structural abnormalities in lumbar tissues, and the clinical course is largely determined by adaptive or maladaptive responses of the nervous system involved in pain processing [2]. Accordingly, cNSLBP is no longer regarded solely as a biomechanical disorder but rather as a complex condition involving multidimensional neurophysiological processes.

Numerous neuroimaging and neurophysiological studies have demonstrated that CLBP is associated with significant changes in the structural and functional organization of the central nervous system [2–5]. As a consequence of these changes, not only the processing of painful stimuli but also the processing of peripheral sensory inputs is altered, leading to a reduced capacity to discriminate between sensory signals and to atypical perception of otherwise innocuous stimuli [6]. Consistent with this notion, reduced pressure pain thresholds, enhanced temporal summation, and diminished tactile spatial acuity have been repeatedly documented in patients with CLBP [7–10].

Within this framework, central sensitization (CS) represents a key phenomenon that warrants particular attention. Present in approximately half of individuals with CLBP, CS has been associated with greater disability and poorer psychological outcomes, and is thought to substantially alter both pain intensity and sensory processing mechanisms [11]. While reduced pressure pain thresholds are consistently observed in CLBP patients with CS compared with those without CS, findings regarding other quantitative sensory modalities remain inconsistent [12–14]. Moreover, there is currently no consensus regarding the clinical implications of sensory processing alterations in the presence of CS.

The International Association for the Study of Pain emphasizes the systematic assessment of somatosensory processing as a fundamental component in distinguishing between nociceptive, neuropathic, and nociplastic pain subtypes in the current pain classification [15]. However, the absence of specific recommendations on which tests should be used to assess these phenomena affords clinicians and researchers considerable methodological flexibility, while simultaneously raising concerns about the classification system’s practical applicability and sensitivity. Indeed, a recent field study reported that expected hypersensitivity findings were not identified in all patients, despite the use of commonly applied sensory assessment methods [16].

The somatosensory temporal discrimination threshold (STDT), which reflects the ability to perceive the temporal interval between two successive stimuli, provides insight into temporal processing, cortical inhibition, and sensory integration [17]. Studies have shown that STDT can be assessed across different body regions and remains relatively stable in healthy adults [18]. Altered STDT has also been linked to chronic pain states characterized by CS [19]. However, evidence regarding STDT in CLBP, particularly within a CS-oriented framework, remains limited. Although Tertemiz and Tepe [20] demonstrated prolonged STDT in individuals with cNSLBP, its interpretation is constrained by several limitations, including the lack of consideration of CS, absence of a biopsychosocial framework, assessment of extremities rather than the lumbar region, and a relatively small sample size.

The primary aim of this study was to compare STDT measured at the lumbar region and dorsum of the hand among patients with cNSLBP with and without CS-related features, and healthy controls. The secondary aim was to examine the relationship between STDT, pain catastrophizing, and disability. The authors hypothesized that chronic pain-related cortical reorganization impairs temporal processing, that this impairment relates to biopsychosocial factors, and that it is more pronounced in individuals with CS-related features. They expected both CLBP groups to have higher STDT values than healthy controls, and for STDT to be significantly associated with clinical outcome measures.

MATERIALS AND METHODS

1. Study design

This single-center, analytical, cross-sectional study included three groups: cNSLBP patients with CS-related features, cNSLBP patients without CS-related features, and healthy controls. The study was conducted at the Department of Physical Medicine and Rehabilitation at Faculty of Medicine, Sivas Cumhuriyet University between August 1, 2021 and July 31, 2022. The study protocol was approved by the Sivas Cumhuriyet University Non-Interventional Research Ethics Committee (decision no: 2021-06/33; June 23, 2021). All participants were informed verbally and in writing, and written informed consent was obtained prior to participation. The study was carried out in accordance with the principles of the Declaration of Helsinki and Good Clinical Practice guidelines.

2. Participants

A total of 111 individuals were included, with 37 per group. Among patients presenting to the clinic, those diagnosed with cNSLBP according to the Turkish consensus report were consecutively recruited [21]. Healthy controls were recruited from hospital and university staff aged 18–65 years with no history of musculoskeletal pain and were matched to the cNSLBP participants by age and sex.

Common inclusion criteria for both cNSLBP patient groups were: age between 18 and 65 years; low back pain persisting for at least 3 months, absence of a specific pathology explaining the pain based on clinical, radiological, and laboratory evaluations; and an average pain intensity of ≥ 3 cm on a 0–10 cm visual analog scale (VAS) during the last week. Exclusion criteria included radicular pain; positive neurodynamic tests on physical examination; presence of motor or reflex deficits; history of lumbar spine surgery; peripheral neuropathy or other neurological or metabolic disorders that could affect somatosensory function (e.g., multiple sclerosis, epilepsy, diabetes mellitus, hypothyroidism); systemic inflammatory rheumatic diseases; a prior clinical diagnosis of fibromyalgia based on medical history and available medical records; active infection, malignancy, or severe cardiopulmonary disease; cognitive impairment or psychiatric disorders; use of medications known to markedly alter central nervous system excitability; receipt of interventional pain treatments (e.g., epidural injections, neural blocks) or structured physical therapy/rehabilitation programs within the previous 6 months; pregnancy; or dermatological conditions affecting skin integrity at the measurement sites. Individuals unable to comply with the assessment procedures were excluded from the study.

The predominance of CS in cNSLBP patients was evaluated according to previously published protocols, and patients were grouped accordingly [22,23]. Pain intensity, distribution, and characteristics were examined. In addition, the Widespread Pain Index (WPI), Symptom Severity Scale (SSS), and Central Sensitivity Inventory (CSI) were evaluated. The WPI evaluates pain presence across 19 anatomical regions over the past week; the SSS assesses core symptoms related to central sensitivity, such as fatigue, headache, and cognitive difficulties [14,22,24]; and the CSI quantifies the severity of CS-related symptoms on a 0–100 scale [14,25,26]. cNSLBP patients meeting the criteria of WPI ≥ 7 and SSS ≥ 5, or WPI between 4–6 with SSS ≥ 9, in combination with a CSI total score ≥ 40, were grouped as patients with CS-related features. To enhance phenotypic characterization of CS and to mitigate the limitations of self-reported assessments, WPI and SSS criteria were combined with the CSI ≥ 40 threshold.

3. Assessment procedures

Demographic characteristics (age, sex), body mass index (BMI), comorbidities, and regular medication use were recorded for all participants using a structured case report form at the time of group allocation. For the cNSLBP groups, pain intensity (VAS, 0–10 cm) and pain duration (months) were also documented.

The primary outcome measure, STDT, was assessed by an experienced evaluator blinded to participants’ sociodemographic characteristics, clinical data, and secondary outcome measures, including the Roland–Morris Disability Questionnaire (RMDQ), Pain Catastrophizing Scale (PCS), and Hospital Anxiety and Depression Scale (HADS). Secondary outcome measures were administered by a blinded physiatrist who was unaware of participants’ sociodemographic characteristics, clinical data, and STDT measurement results. Primary and secondary outcome measures were completed on the same day, in a randomized order, to minimize potential sequence effects. Given their pain-specific nature, RMDQ and PCS were assessed only in the cNSLBP groups.

The RMDQ is a 24-item yes/no self-report questionnaire assessing low back pain-related functional limitation, with higher scores indicating greater disability [27,28]. Pain catastrophizing was assessed using the 13-item PCS, which evaluates catastrophic thinking related to pain, with higher scores reflecting greater catastrophizing [29,30]. Emotional status was assessed using the HADS, a 14-item self-report scale comprising anxiety (HADS-A) and depression (HADS-D) subscales, with higher scores indicating greater symptom severity [28,31].

4. Somatosensory temporal discrimination threshold assessment procedures

STDT was evaluated based on previously described standardized protocols and adapted to the anatomical sites assessed in the present study [18]. All measurements were performed in a quiet laboratory environment with a controlled ambient temperature (22°C–24°C), with participants lying comfortably in the prone position. Lumbar STDT measurements were obtained at the midpoint of the area identified by each participant with cNSLBP as the site of maximal pain on a lumbar pain diagram. In healthy controls, lumbar STDT was assessed at a standardized anatomical site located 5 cm lateral to the right of the L3 spinous process [32]. Hand measurements were performed on the dorsum of the dominant hand, between the second and third metacarpal bones. To reduce potential order effects, the measurement protocol was applied in a randomized sequence across participants and balanced between groups. Prior to testing, participants underwent a brief familiarization session during which the procedure was explained and comprehension of instructions was ensured.

Before assessment, the skin surface was cleaned with an antiseptic solution. Anodal and cathodal surface electrodes with a 10-mm diameter were placed 1 cm apart and securely fastened. Tactile stimulation was delivered using paired square-wave electrical pulses of 0.2 ms duration, controlled by a constant-current stimulator (Keypoint; Medtronic). To determine the individual sensory perception threshold, stimulation was initiated at 1 mA and increased in 0.2 mA increments until the stimulus was consistently perceived. This procedure was repeated three times, and the lowest intensity consistently perceived across all trials was recorded as the sensory threshold. The STDT assessment was then performed with a stimulus intensity set at 1.5 times the individual's sensory threshold.

The interstimulus interval (ISI) between paired stimuli was initially set at 0 ms and increased in 5 ms increments. Paired stimuli were delivered every 10 seconds, and participants were asked to report whether they perceived one or two stimuli. The smallest ISI at which two stimuli were correctly identified in three consecutive trials was recorded as the ascending STDT. Subsequently, the ISI was decreased in 5 ms steps, and the first ISI at which participants reported perceiving a single stimulus in three consecutive trials was recorded as the descending STDT. Ascending and descending STDT measurements were repeated twice for each anatomical site, and the arithmetic mean of these values was calculated as the final STDT for that site. Separate STDT values were obtained for the dorsum of the hand and the lumbar region for each participant.

5. Statistical analysis

Sample size estimation was based on the primary hypothesis that STDT values would differ among the three groups (cNSLBP with CS-related features, cNSLBP without CS-related features, and healthy controls). Using G*Power software (version 3.1.9.7), a one-way analysis of variance (ANOVA) comparing three independent groups was planned with a significance level of α = 0.05 and a statistical power of 1−β = 0.80. Although Tertemiz and Tepe [20] reported a large effect size (Cohen’s d = 2.3) for STDT differences between patients with CLBP and healthy controls, a more conservative approach was adopted in the present study due to the inclusion of three groups and assessment of both lumbar and extremity STDT. Accordingly, a medium effect size (f = 0.30) was assumed, resulting in a required minimum total sample size of 111 participants.

Statistical analyses were performed using IBM SPSS Statistics (version 23; IBM Co.). A P value < 0.05 was considered statistically significant. Normality of continuous variables was assessed using histograms, Q–Q plots, and the Shapiro–Wilk test. Continuous variables are presented as mean ± standard deviation, whereas categorical variables are presented as counts and percentages.

Between-group comparisons of sociodemographic and clinical variables were conducted using one-way ANOVA for continuous variables and the chi-square test for categorical variables. Post hoc pairwise comparisons following ANOVA were performed using Bonferroni correction.

Hand and lumbar STDT values were initially compared among the three groups using one-way ANOVA. Subsequently, between-group differences were further examined using a general linear model (GLM) approach with potential confounding variables included as covariates, and effect sizes were reported as partial eta squared (partial η²). Two separate models were constructed. In the first model, covariates common to all three groups—age, sex, years of education, BMI, presence of comorbidities, and total HADS score—were included. Pairwise group comparisons were evaluated using Bonferroni-adjusted estimated marginal means.

The second model was constructed to compare the cNSLBP with CS-related features group and the cNSLBP without CS-related features group. In this model, symptom duration, pain intensity, and PCS score were included as covariates. Due to multicollinearity between PCS and the RMDQ score, the latter was not included in the model; instead, PCS was selected as the covariate representing pain-related cognitive processes more directly associated with CS.

Associations between hand and lumbar STDT values and pain catastrophizing and functional disability were examined using Pearson correlation analyses conducted separately for each group. Group comparisons of STDT, the primary outcome measure, were treated as confirmatory analyses, whereas covariate-adjusted models and within-group correlation analyses were considered exploratory. Bonferroni correction was used for post hoc pairwise group comparisons, whereas no additional formal multiplicity correction was applied to the exploratory analyses. Given the limited number of prespecified exploratory tests, emphasis was placed on effect sizes and the magnitude of associations rather than on P values alone.

As an additional exploratory analysis, a residualized STDT-based unsupervised clustering analysis was performed to examine whether covariate-adjusted STDT measures formed data-driven sensory profiles among patients with cNSLBP. This analysis was restricted to the two cNSLBP groups because the aim was not to distinguish patients from healthy controls, but to evaluate whether adjusted STDT-based sensory profiles corresponded to the presence or absence of CS-related clinical features among patients.

Before clustering, hand and lumbar STDT values were residualized using separate linear regression analyses adjusted for age, sex, years of education, BMI, presence of comorbidities, VAS, and symptom duration. These variables were selected as demographic and clinical covariates that may influence STDT values and clinical heterogeneity among patients with cNSLBP. PCS, HADS-D, HADS-A, RMDQ, and variables used to define CS-related features were not included in the residualization analyses to avoid overadjustment or circularity, because these variables were considered part of, or closely related to, the clinical CS-related phenotype rather than core non-phenotypic confounders. Multicollinearity among the covariates included in the residualization analyses was assessed using variance inflation factors. The distribution of standardized residuals was inspected, and potential outliers were checked before clustering because Euclidean distance-based methods may be sensitive to extreme values. Predefined clinical group labels were not included in the residualization analyses or during clustering. The standardized residuals obtained from the hand and lumbar STDT regression analyses were then entered into an agglomerative hierarchical clustering analysis.

Squared Euclidean distance was used as the distance metric because both clustering variables were continuous residualized STDT measures and because it is consistent with Ward’s minimum variance method, which minimizes the within-cluster sum of squares. Ward’s minimum variance method was used as the linkage method to obtain compact sensory profiles. A two-cluster solution was prespecified because the analysis aimed to examine the correspondence between adjusted STDT-based sensory profiles and the two clinically predefined cNSLBP phenotypes, namely patients with and without CS-related features. The dendrogram was used to visualize the hierarchical clustering structure. To maintain consistency with the prespecified clinical hypothesis, the dendrogram was cut at the level corresponding to a two-cluster solution, rather than determining the number of clusters post hoc.

Cluster quality was assessed using the silhouette coefficient, which ranges from −1 to 1, with higher values indicating better within-cluster cohesion and between-cluster separation. Agreement between the residualized STDT-based clusters and the predefined clinical groups was evaluated descriptively using cross-tabulation and the adjusted Rand index. The resulting clusters were labeled according to their mean residualized STDT values as lower residualized-STDT and higher residualized-STDT sensory profiles. This analysis was considered exploratory and hypothesis-generating and was not intended to develop a diagnostic or predictive model. The exploratory residualization and clustering analyses were performed using Python version 3.13.5.

RESULTS

In this study, which evaluated 181 individuals, 23 individuals did not meet the inclusion criteria, 35 individuals had one or more exclusion criteria, and 12 individuals (5 from the cNSLBP group with CS-related features and 7 from the cNSLBP group without CS-related features) were excluded from the study because they could not comply with the STDT measurement procedure. Results from 111 participants, with 37 per group, were analyzed (Fig. 1).

Fig. 1.

Flow diagram of participant selection. cNSLBP: chronic non-specific low back pain, CS: central sensitization, STDT: somatosensory temporal discrimination threshold.

The sociodemographic data of the three groups were statistically similar (Table 1). The HADS-D and HADS-A scores of cNSLBP patients with CS-related features were higher than those of other groups (all P < 0.001).

Table 1.

Demographic and clinical characteristics of the study groups

	Healthy controls (n = 37)	Patients without CS-related features (n = 37)	Patients with CS-related features (n = 37)	P value
Age (yr)	51.9 ± 5.2	51.3 ± 5.1	52.5 ± 5.1	0.606
BMI (kg/m2)	26.2 ± 0.9	26.4 ± 1.0	26.4 ± 0.9	0.588
Education duration (yr)	10.0 ± 3.2	10.1 ± 3.4	10.0 ± 3.1	0.946
Sex, female	24 (64.9)	24 (64.9)	24 (64.9)	> 0.999
Comorbidity				0.604
Hypertension	30 (81.1)	30 (81.1)	26 (70.3)
Asthma	2 (5.4)	2 (5.4)	3 (8.1)
Coronary artery disease	1 (2.7)	1 (2.7)	3 (8.1)
GERD	2 (5.4)	2 (5.4)	2 (5.4)
Nephrolithiasis	2 (5.4)	2 (5.4)	3 (8.1)
HADS-depression	2.6 ± 1.2	9.9 ± 2.4	13.7 ± 1.1	< 0.001
HADS-anxiety	2.3 ± 1.2	8.9 ± 2.5	12.8 ± 1.4	< 0.001

Values are presented as mean ± standard deviation or number (%). One-way ANOVA was used for continuous variables. Categorical variables were compared using the chi-square test. Bonferroni correction was applied for post-hoc pairwise comparisons.

CS: central sensitization, BMI: body mass index, GERD: gastroesophageal reflux disease, HADS: Hospital Anxiety and Depression Scale.

The mean pain duration was similar between two cNSLBP groups (P = 0.845). cNSLBP patients with CS-related features had higher pain intensity, higher PCS, and higher RMDQ scores (P = 0.035, P < 0.001, P < 0.001, respectively) (Table 2).

Table 2.

Pain-related clinical characteristics of patients with cNSLBP

	Patients without CS-related features (n = 37)	Patients with CS-related features (n = 37)	P value
Pain intensity (VAS, 0–10 cm)	5.1 ± 1.1	5.7 ± 1.1	0.035
Symptom duration (mo)	26.3 ± 9.8	25.9 ± 10.3	0.845
PCS	14.4 ± 2.1	34.0 ± 3.8	< 0.001
RMDQ	12.8 ± 1.5	15.8 ± 1.6	< 0.001

Values are presented as mean ± standard deviation. Independent samples t-test was used for analysis.

cNSLBP: chronic non-specific low back pain, CS: central sensitization, VAS: visual analogue scale, PCS: Pain Catastrophizing Scale, RMDQ: Roland–Morris Disability Questionnaire.

Mean hand STDT durations were 55.43 ± 10.46 ms in the cNSLBP patients with CS-related features, 35.76 ± 4.51 ms in the cNSLBP patients without CS-related features, and 32.86 ± 4.03 ms in the healthy group (Fig. 2). One-way ANOVA showed a significant difference in hand STDT durations between groups (F[2,108] = 114.69; P < 0.001). Post-hoc analyses revealed that cNSLBP patients with CS-related features had significantly longer hand STDT duration compared to both the healthy group (mean difference = 22.57; P < 0.001) and the cNSLBP patients without CS-related features (mean difference = 19.68; P < 0.001). No significant difference was observed between the cNSLBP patients without CS-related features and the healthy group (P = 0.232). In the covariance-adjusted GLM analysis, the group effect on hand STDT duration remained significant (F[2,101] = 50.20; P < 0.001; partial η² = 0.499). Furthermore, in the pairwise comparison between the two cNSLBP groups, taking clinical variables into account, the group effect was also found to be significant (F[1,69] = 32.55; P < 0.001; partial η² = 0.321).

Fig. 2.

Between-group comparisons of hand and lumbar STDT. Boxes represent the interquartile range, center lines the median, and whiskers extend to the most extreme values. (A) Hand STDT and (B) lumbar STDT. STDT: somatosensory temporal discrimination threshold, CS: central sensitization.

Mean lumbar STDT durations were 101.95 ± 7.09 ms in cNSLBP patients with CS-related features, 89.73 ± 7.47 ms in cNSLBP patients without CS-related features, and 65.81 ± 10.44 ms in the healthy group (Fig. 2). One-way ANOVA demonstrated a significant difference in lumbar STDT durations between groups (F[2,108] = 174.36; P < 0.001). Post-hoc analyses revealed that cNSLBP patients with CS-related features had significantly longer lumbar STDT durations compared to both the healthy group (mean difference = 36.14 ms; P < 0.001) and the cNSLBP patients without CS-related features (mean difference = 12.22 ms; P < 0.001). In addition, the cNSLBP patients without CS-related features exhibited significantly longer lumbar STDT durations than the healthy group (mean difference = 23.92 ms; P < 0.001). In the covariance-adjusted GLM analysis, the group effect on lumbar STDT duration remained significant (F[2,101] = 33.80; P < 0.001; partial η² = 0.401). Furthermore, in the pairwise comparison between the two cNSLBP groups, after adjusting for clinical variables, the group effect was also found to be significant (F[1,69] = 12.52; P = 0.001; partial η² = 0.154).

No significant relationship was found between hand STDT duration and PCS in the cNSLBP patients without CS-related features (r = 0.126; P = 0.459). In the cNSLBP patients with CS-related features, a positive and significant relationship was found between hand STDT duration and PCS (r = 0.330; P = 0.046). No significant relationship was observed between lumbar STDT duration and PCS in the cNSLBP patients without CS-related features (r = 0.275; P = 0.100), while a positive and significant relationship was found in the cNSLBP patients with CS-related features (r = 0.529; P = 0.001) (Fig. 3).

Fig. 3.

Associations of STDT with PCS and RMDQ according to CS-related features. (A) Hand STDT and PCS, (B) lumbar STDT and PCS, and (C) lumbar STDT and RMDQ. STDT: somatosensory temporal discrimination threshold, PCS: Pain Catastrophizing Scale, RMDQ: Roland–Morris Disability Questionnaire, CS: central sensitization.

The relationship between hand STDT duration and the RMDQ was not statistically significant in either group (cNSLBP patients without CS-related features: r = 0.184, P = 0.276; cNSLBP patients with CS-related features: r = 0.200, P = 0.236). The relationship between lumbar STDT duration and the RMDQ index was not significant in the cNSLBP patients without CS-related features (r = 0.085; P = 0.615), while a positive and significant relationship was found in the cNSLBP patients with CS-related features (r = 0.343; P = 0.038) (Fig. 3).

The residualized STDT-based clustering generated two sensory profiles, as illustrated by the dendrogram and scatter plot in Fig. 4. The lower residualized-STDT sensory profile included all patients without CS-related features and 10 patients with CS-related features, whereas the higher residualized-STDT sensory profile included 27 patients with CS-related features and no patients without CS-related features. The overall agreement between the residualized STDT-based cluster allocation and the predefined clinical grouping was 86.5%. The adjusted Rand index was 0.526, indicating moderate agreement beyond chance, and the silhouette coefficient was 0.568, suggesting a moderate degree of cluster cohesion and separation.

Fig. 4.

Hierarchical clustering dendrogram (A) and scatter plot (B) of residualized STDT profiles. cNSLBP: chronic non-specific low back pain, STDT: somatosensory temporal discrimination threshold, CS: central sensitization.

DISCUSSION

In this study, the authors compared lumbar and hand STDT among individuals with cNSLBP with and without CS-related features and healthy controls, and further examined the associations between STDT, pain catastrophizing, and disability. Their findings demonstrate that cNSLBP patients with CS-related features exhibited significantly prolonged STDT values at both the lumbar and hand regions. Moreover, lumbar STDT was associated with both pain catastrophizing and disability, whereas hand STDT showed an association only with pain catastrophizing. To the authors’ knowledge, this is the first study to investigate STDT in cNSLBP while explicitly accounting for the CS-related features.

STDT is regarded as a neurophysiological measure that reflects the temporal resolution of the central nervous system and the precision of inhibitory mechanisms within the somatosensory cortex [17,33]. Although the primary somatosensory cortex constitutes the core neural substrate, subcortical structures such as the basal ganglia and thalamus also contribute to this distributed temporal processing network [17]. In contrast to tactile acuity measures, which predominantly assess the spatial organization, STDT probes the temporal precision of cortical processing by evaluating the ability to segregate successive non-nociceptive tactile stimuli over time [34,35]. While quantitative sensory tests such as pressure pain threshold, temporal summation of pain, and conditioned pain modulation (CPM) primarily interrogate nociceptive pathways [36], STDT was specifically selected in the present study to capture alterations in temporal sensory processing and inhibitory control that may underlie CS beyond spatial sensory distortions.

In healthy individuals, STDT values measured from the dorsum of the hand are typically reported to range between approximately 30 and 50 ms [18]. In contrast, prolonged STDT has been observed in various chronic pain conditions. Tertemiz and Tepe [20] similarly reported prolonged STDT values in individuals with CLBP, consistent with the present findings. However, in their study, the measurements were obtained from the extremities rather than the lumbar region. Similarly, prolonged STDT has also been reported in chronic pain-related conditions such as fibromyalgia and rheumatoid arthritis, in which STDT values were found to be significantly higher than those in healthy controls [19,37]. Taken together, these clinical observations may be consistent with the biological plausibility of altered central somatosensory processing in chronic pain conditions. An updated review of the STDT literature indicates that somatosensory temporal discrimination depends on intracortical inhibitory processing within S1 and on broader basal ganglia-thalamocortical interactions [17]. Accordingly, the elevated lumbar STDT observed in this study’s cNSLBP groups may be interpreted as being consistent with altered central somatosensory processing, although the present study does not directly demonstrate the underlying neural mechanism. Previously, reduced tactile acuity has been linked to structural neuroplasticity in primary somatosensory cortex, and complementary structural and resting-state magnetic resonance imaging (MRI) data have shown altered S1 connectivity together with morphometric differences in the S1 region representing the low back [34].

CS is characterized by enhanced synaptic plasticity and insufficient inhibitory control, leading to neurophysiological changes that affect not only nociceptive pathways but also non-nociceptive sensory processing systems [38]. Accordingly, CS is increasingly conceptualized as a global disturbance of sensory modulation rather than a purely pain-specific hypersensitivity [39]. This framework aligns with the concept of multimodal hypersensitivity frequently described in chronic pain syndromes, in which the central nervous system exhibits exaggerated or dysregulated processing across tactile, visual, and auditory modalities [39]. Consistent with this view, conditions in which CS predominates have demonstrated impaired sensory gating even in response to non-nociceptive stimuli, suggesting that dysfunction within nociceptive systems may extend to lemniscal pathways and broader inhibitory circuits [40]. Importantly, these sensory processing deficits are not confined to the primary pain location but instead reflect generalized central nervous system involvement. Central neuroplastic changes are known to transcend dermatomal boundaries, allowing sensory abnormalities to spread beyond the primary pain area and manifest as non-dermatomal somatosensory deficits [11]. This interpretation is further supported by human neuroimaging findings in chronic pain. In CLBP, functional MRI during non-painful lumbar mechanosensory stimulation has demonstrated reorganization in the secondary somatosensory cortex, while resting-state functional MRI has shown that chronic pain pathophysiology and clinical pain intensity are associated with abnormal thalamocortical network dynamics involving the ventral lateral/posterolateral thalamus, postcentral gyrus, and insular regions [41]. In addition, a systematic review of CLBP neuroimaging studies found moderate evidence for regional gray and white matter changes, altered functional connectivity during rest, and increased activity in pain-related brain regions [5]. In line with this framework, the present findings of prolonged STDT not only at the lumbar region but also at the dorsum of the hand in cNSLBP participants with CS-related features may reflect the presence of widespread alterations in temporal sensory processing. Indeed, Santiesteban et al. [42] reported poorer tactile acuity in individuals with non-specific persistent low back pain and predominant CS than in those with CLBP without CS.

In addition, both hand and lumbar STDT values were associated with PCS in the cNSLBP with CS-related features. This pattern may be associated with a meaningful interaction between cognitive-emotional processes and sensory-temporal processing. Pain catastrophizing, characterized by heightened threat perception, ruminative thinking, and feelings of helplessness, has been linked to impaired top-down inhibitory pain modulation and reduced effectiveness of descending control systems [43]. Within this framework, catastrophizing is increasingly viewed not merely as an emotional response but as a factor associated with measurable physiological alterations in central pain processing. Conversely, CS itself—through increased central gain and compromised inhibitory control—may create a neurophysiological milieu that sustains catastrophic cognitive schemas, giving rise to a bidirectional relationship between these processes [11]. This interaction is consistent with the frequently reported combination of impaired CPM responses and heightened central pain gain in chronic pain states dominated by CS [44]. Given the proposed link between STDT and central inhibitory and temporal processing networks, the observed association between STDT and PCS scores may be interpreted within this integrative model. Notably, the association between PCS and hand STDT may be consistent with the notion that sensory processing disturbances in CS are not limited to the primary pain region, but may also involve more widespread supraspinal network involvement.

In the group with CS-related features, lumbar STDT values were also associated with RMDQ. Disability is a multidimensional outcome shaped not only by pain intensity but also by physical capacity, avoidance behaviors, psychosocial context, and emotional state [45]. This association may suggest that as phenotypic severity increases, sensory processing impairments become more closely linked to clinically meaningful functional limitations. One possible interpretation is that altered central sensory-sensorimotor integration contributes to the observed relationship between lumbar STDT and disability. Neuroimaging studies in CLBP have reported abnormalities involving S1, S2, and thalamocortical networks, supporting the possibility that altered central sensory processing may contribute to functionally relevant sensorimotor disturbances [46,47]. Within this framework, impaired integration of proprioceptive and temporal information from the trunk may contribute to sensorimotor incongruence and may be related to protective muscle activation, movement avoidance, and rigid motor control strategies [48,49]. Together, these processes may form a self-perpetuating cycle that may contribute to the progression of functional disability. However, because no neuroimaging or neurophysiological recordings were obtained in the present cohort, this explanation should be regarded as indirect and hypothesis-generating rather than confirmatory.

Several limitations of this study should be acknowledged. A major limitation is that no direct neurophysiological or neuroimaging assessments were obtained. Although STDT provides a psychophysical measure of somatosensory temporal processing, the absence of electroencephalography, functional MRI, or other objective measures of cortical, thalamocortical, or spinal processing precludes any confirmatory interpretation regarding the neural mechanisms underlying the observed STDT alterations. Therefore, the present findings should be interpreted as behavioral and clinical associations consistent with altered central sensory processing, rather than as direct evidence that CS caused the STDT changes. Future studies combining STDT with electroencephalography, functional MRI, quantitative sensory testing, or other objective neurophysiological methods are needed to confirm these mechanisms. Second, only one sensory modality was evaluated; this does not allow for comparison with other modalities. Third, the cross-sectional design precludes causal inference, and findings should be interpreted as associative. Fourth, the classification of CS-related features relied on questionnaire-based and self-reported measures, which may have introduced some degree of misclassification. Fifth, lumbar STDT measurements were obtained from the site of maximal pain in participants with cNSLBP, whereas a predefined standardized anatomical site was used in healthy controls; although this approach was intended to improve feasibility and standardization, the possible influence of lumbar test-site selection on between-group lumbar STDT comparisons should be considered when interpreting these findings. Also, the single-center design may also limit generalizability to broader clinical and sociodemographic populations. Finally, although several demographic and clinical variables were assessed, residual confounding cannot be fully excluded. In particular, some causes and risk factors potentially relevant to cNSLBP were not comprehensively evaluated, including physical activity level, smoking status, occupational physical loading, ergonomic exposures, and psychosocial or socioeconomic characteristics. These unmeasured factors may have influenced both group classification and STDT-related outcomes. In addition, because multiple exploratory correlation analyses were performed, the possibility of type I error should also be considered when interpreting these findings.

In conclusion, patients with cNSLBP and CS-related features exhibit prolonged STDT at both the lumbar region and the dorsum of the hand compared with those without CS-related features and healthy controls. Lumbar STDT was associated with both pain catastrophizing and disability, whereas hand STDT was associated exclusively with pain catastrophizing. Overall, the authors’ findings may suggest the view that sensory alterations in cNSLBP with CS-related features may extend beyond the local pain area and may reflect a more widespread disturbance in central sensory processing. These findings may also suggest that the clinical expression of sensory-temporal processing abnormalities in CS may vary according to anatomical region. However, these findings should be interpreted cautiously given the cross-sectional design, the possibility of residual confounding from unmeasured factors, and the absence of direct neuroimaging or neurophysiological measurements. Future longitudinal and prospective studies are warranted to clarify the temporal dynamics and directionality of the relationship between CS and STDT. Integrating STDT with other quantitative sensory tests, such as pressure pain threshold, temporal summation, and CPM, may enable a more comprehensive characterization of the sensory phenotype of CS. Moreover, combining STDT assessments with neuroimaging or neurophysiological techniques could provide valuable insights into the cortical and subcortical mechanisms underlying altered temporal sensory processing. Finally, multicenter studies across diverse chronic pain populations are needed to determine the clinical utility and generalizability of STDT as a marker of CS-related sensory dysfunction.