Fatigue-induced biomechanical decoupling at L4-S1 discs: mechanism of disc degeneration in chronic low back pain

Abstract

Objective

This study aims to quantitatively compare segment-specific disc kinematics, height changes, and strain patterns in the lumbar spine between patients with chronic low back pain (CLBP) and asymptomatic controls during weight-bearing flexion-extension, before and after fatigue loading.

Methods

A total of 29 patients with CLBP and 24 asymptomatic controls were included. All participants underwent dual fluoroscopic imaging system (DFIS) and computed tomography (CT) imaging. Weight-bearing kinematic sequences from maximum extension to approximately 45° flexion were captured pre- and post-fatigue. A validated geometric deformation model based on supine CT scans was used to quantify intervertebral kinematics from L1 to S1. Disc height variations and strain (axial and shear) were computed using computational morphometry and kinematic correspondence tracking. Statistical analyses included paired and independent t-tests for within- and between-group comparisons, with significance set at p < 0.05.

Results

Computational morphometry revealed distinct degenerative features associated with CLBP. Compared with controls, L4/5 discs in CLBP patients showed significantly reduced baseline height (p < 0.05), which may indicate a predisposition to early degeneration. At L5/S1, disc height was significantly reduced after fatigue loading across all postures in the CLBP group (p < 0.01), a change not observed in controls. Strain analysis demonstrated segment-dependent alterations in mechanical behavior. During flexion, both groups exhibited posterior tensile and anterior compressive strains from L4 to S1. However, after fatigue, CLBP patients developed a distinct strain pattern at L5/S1, characterized by high compressive strain and reduced shear strain. More broadly, fatigue loading in CLBP patients led to impaired shear strain dissipation and accumulation of compressive strain within the L4–S1 segments. In contrast, the upper lumbar segments (L1–L3) in both groups maintained physiological strain gradients that were unaffected by fatigue.

Conclusions

This study provides in vivo evidence of fatigue-induced biomechanical alterations in the L4–S1 discs of CLBP patients, marked by increased compressive strain and reduced capacity for shear strain redistribution. The preserved biomechanical behavior in the upper lumbar segments underscores the particular vulnerability of the lumbosacral junction to mechanical deficits. These findings support the value of segment-specific biomechanical assessment for identifying early degenerative changes and inform the development of targeted load-management strategies in CLBP. However, further research is needed to account for potential confounding variables and to validate these observations in larger cohorts.

Trial registration

Chinese Clinical Trial Registry Registration number ChiCTR2000036039. Registered 21 August 2020.

Background

Chronic low back pain (CLBP), characterized by persistent pain in the lumbar region lasting beyond three months, represents a major global health burden affecting individuals across all age groups [1, 2]. With a lifetime prevalence of 84% [3], CLBP is the leading cause of disability worldwide, significantly contributing to work absenteeism and premature retirement due to disability [4]. Its etiology extends beyond musculoskeletal impairments—such as reduced strength, function, and mobility—to encompass complex physiological mechanisms and modern lifestyle factors [5].

Fatigue, a prevalent physiological phenomenon, is widely recognized as a significant contributor to CLBP. It not only induces or exacerbates symptoms but also compromises lumbar function, particularly the biomechanical behavior of intervertebral segments [6, 7]. As the most caudal mobile section of the human spine, the lumbar spine plays a critical role in supporting the upper body by transmitting forces and bending moments to the sacrum [8, 9]. Degenerative changes within lumbar discs can lead to loss of structural integrity and subsequent intractable CLBP [10]. While various biological factors contribute to disc pathology [11–14], aberrant spinal motion and the resulting abnormal loading are widely regarded as primary etiological factors in intervertebral disc degeneration [7, 15]. Although clinical and imaging studies consistently link lumbar disc degeneration with CLBP, the precise in vivo, segment-specific biomechanical mechanisms—particularly the dynamic deformation of discs under physiological load—remain poorly understood. A critical gap exists in quantifying how these deformations vary between spinal levels and are influenced by states of muscle fatigue, which our study aims to address.

Studies indicate that prolonged poor postural habits (e.g., sustained sitting, forward bending during lifting) accelerate both disc degeneration and muscular strain [16–18]. Furthermore, paraspinal muscle fatigue and erector spinae weakness undermine spinal stability, establishing a detrimental “muscle-disc” vicious cycle [19]: muscular fatigue induces compensatory postures that increase lumbar pressure, while disc deformation heightens muscular tension and inflammatory responses [20, 21]. Therefore, a comprehensive understanding of spinal biomechanics is essential for elucidating the pathogenesis of lumbar disc disorders and developing therapeutic strategies aimed at restoring lumbopelvic function.

Previous studies employing direct measurement techniques provided limited but valuable experimental data regarding static intradiscal pressure distribution [22, 23] and spinal load estimation [22]. However, these investigations present significant limitations: in vitro models fail to replicate tissue complexity in vivo, and invasive puncture techniques, which carry risks of iatrogenic degeneration, are no longer recommended. Advancements in non-invasive imaging modalities now enable in vivo monitoring of load-bearing tissue biomechanics.

This study utilizes dual fluoroscopic imaging system (DFIS) to capture three-dimensional skeletal kinematics of the lumbopelvic region during functional movements, both pre- and post-fatigue induction. Combined with individualized vertebral morphological parameters derived from CT scans, this approach provides a unique opportunity to investigate in vivo morphological characteristics and dynamic deformation of the L1-S1 intervertebral discs. Leveraging this technical platform, we systematically quantified compression and shear deformation at adjacent L1-S1 segments, evaluating deformation characteristics and patterns in CLBP patients during functional movements from maximum extension to approximately 45° flexion, both before and after fatigue induction. Our hypothesis posits that: (1) Lumbar disc deformation during physiological weight-bearing flexion-extension demonstrates segment-specific characteristics, and (2) The magnitude of deformation significantly correlates with fatigue state.

Methods

Participant recruitment

This study received prior ethical approval from the Institutional Review Board of Tongji Hospital, Shanghai, China (Protocol No. 2021-011-SK) and strictly adhered to the ethical principles outlined in the Declaration of Helsinki [24]. Participant recruitment occurred through investigators’ personal and professional networks. Written informed consent was obtained from all participants prior to data collection, with emphasis placed on the confidentiality of personal health information.

Inclusion criteria comprised: a clinical diagnosis of chronic low back pain (CLBP), defined as pain localized between the twelfth rib and the inferior gluteal folds, with or without referred leg pain, lasting for more than 12 weeks, age between 18 and 40 years; ability to perform standardized lifting and sagittal plane bending maneuvers; and native Chinese language proficiency to ensure protocol comprehension. Exclusion Criteria were any of the following: (1) Structural spinal abnormalities; (2) Lower extremity referred pain extending beyond the gluteal fold;3) Prior surgical intervention involving spinal, hip, or knee joints; 4) Radicular pain or neurological deficits (motor weakness, sensory disturbances); 5) History of spinal neoplasms or infections; 6) Uncorrected visual or auditory deficits. An age-matched healthy control (HC) group (18–40 years) with no history of spinal symptoms was recruited using the same selection protocol. Additional exclusion criteria for HC participants were: (1) back pain at the time of testing, (2) an episode of back pain that had necessitated attending a medical practitioner or allied health professional in the last 12 months, (3) time off work due to back pain in the last 12 months or, (4) any back pain during or between testing procedures. Initial screening was performed by a board-certified orthopedic surgeon, with final eligibility confirmed by a senior physical therapist. All CLBP participants underwent clinical evaluation using standardized self-reported measures: the Visual Analog Scale (VAS) [25] for pain intensity and Modified Oswestry Disability Index (ODI) for functional limitation [26, 27].

Lumbopelvic 3D reconstruction

All participants underwent lumbopelvic CT scans covering vertebral levels L1 to S1 in a standardized supine position. Imaging was performed using a United Imaging uCT760 scanner (Shanghai United Imaging Healthcare Co., Ltd., China) with the following parameters: 140 kvA and a 512 × 512 matrix. Image reconstruction was conducted at a slice interval of 1.0 mm. The resulting CT datasets were imported into Amira 6.7 software (Thermo Fisher Scientific, Rockford, IL, USA) for three-dimensional bone model reconstruction, thereby establishing precise anatomical references for subsequent kinematic analyses.

Kinematic acquisition protocol

Spinal kinematics were dynamically captured using a DFIS (TAOiMAGE Biplane X-Ray System, Shanghai, China; 2804 × 2804pixel resolution). Participants performed weight-bearing forward-backward bending from maximum trunk extension to approximately 45° of flexion (measured using a protractor) to keep the participants within view of the system [28]. Standardized positioning included: (1) initial palm placement on the anterior thighs with guided motion along marker trajectories, (2) Images were acquired while participants maintained each predefined posture with hands clasped behind the head for approximately 2 s, and (3) maintenance of 45° oblique lower limb alignment to enhance visualization of bony structures (Fig. 1). Pre-experimental training ensured movement consistency; deviations from the protocol resulted in trial repetition. Lead shielding was applied to non-imaged regions in compliance with radiation safety standards.

Fig. 1.

Synchronized biplane radiography captured lumbopelvic (L1-S1) motion during participant trials from maximum extension to ~ 45° flexion

The experimental sequence consisted of three phases:1) Baseline Kinematic Assessment: Initial biplanar radiography was performed to capture full flexion–extension cycles under controlled loading conditions.2) Fatigue Induction: Participants performed repetitive stoop-lifting tasks in accordance with our previously validated fatigue protocol [29]. The protocol was continued until a perceived exertion level of Borg scale 17 (“very difficult”) was reached [30]. The number of repetitions required to attain Borg 17 ranged between 15 and 45 across participants. All participants were instructed to consistently use the stoop-lifting technique (trunk flexion with knee extension) throughout the fatigue induction phase. Each full lift-and-lower cycle averaged 8.0 s, and the entire fatigue induction phase was completed within 2–6 min, depending on individual endurance. Real-time monitoring was implemented throughout to ensure participant safety and protocol adherence.3) Post-Fatigue Kinematic Reassessment: Repeat biplanar imaging was conducted under identical conditions as the baseline assessment.

During all lifting trials, participants lifted a 5 kg box with handles. The box had to be grabbed with both hands, lifted up with the elbows extended or slightly flexed (height of the handles in upright standing position: about hip/pelvis height), and placed back on the same place. Heel contact and consistent initial hand placement were maintained throughout. Box handle height (standardized to patellar height level) and foot placement (15 cm posterior to the load) were controlled according to established ergonomic guidelines [31, 32]. No additional constraints were imposed to preserve natural spinal kinematics.

Data processing and analysis

Three-dimensional(3D) L1-S1 configuration under weight-bearing flexion-extension

A segment-specific anatomical coordinate system for L1-S1 vertebrae was established following International Society of Biomechanics recommendations [33]. Biplanar digital radiographs and CT-derived 3D bone models were co-registered using a custom MATLAB algorithm (R2022b; MathWorks, Natick, MA, USA), iteratively optimizing each vertebra’s spatial orientation until achieving optimal contour matching between projected DR silhouettes and model geometries [29]. This technique has been validated against Roentgen stereophotogrammetric analysis (RSA)—the gold standard for in vivo skeletal motion tracking—demonstrating submillimeter accuracy [34, 35].

Intervertebral disc height measurement

Disc height was quantified based on established computational morphometric methods [36]. Using CT-based 3D models, approximately 3,000 uniformly distributed points were generated across the surfaces of adjacent vertebral endplates. Spatial coordinates were obtained from the registered weight-bearing models, and the minimum Euclidean distances between corresponding superior and inferior endplate point clouds were computed using an iterative closest point (ICP) algorithm. The arithmetic mean of these distances was defined as the segmental disc height (Fig. 2a).

Fig. 2.

Calculation of Disc Height and Strain a Intervertebral disc height was measured as the minimum distance between adjacent endplates in intrinsic space using the Iterative Closest Point (ICP) algorithm. b Spatial projection of the minimum-distance vector onto X-, Y-, and Z-axes. Regional disc strain was computed from projection variations during flexion-extension motion: 1) Axial strain: Derived from length changes along Z-axis (ΔA); 2) Shear strain: Calculated from XY-plane deformation (ΔBC)

Disc strain quantification

Disc strain was evaluated to characterize distribution patterns and trends of deformation under different weight-bearing postures. Using supine CT scans as non-weight-bearing references, a geometric deformation model was developed within MATLAB to assess intervertebral kinematics from L1/L2 to L5/S1. Each disc was represented as a triangular mesh with endplate correspondence established via ICP registration. Three-dimensional displacement vectors between endplate coordinate systems quantified deformation magnitude under each loaded posture (Fig. 2b). The strain computation, based on these validated kinematic outputs, was performed using custom algorithms in MATLAB. Strain analysis incorporated the following components [37]:

• Axial strain: Computed as normalized z-axis displacement (superior-inferior direction) relative to supine disc height, with positive/negative values indicating tensile/compressive strain, respectively.

• Shear strain: Defined as normalized in-plane (x-y) displacement relative to supine disc height.

Given the complex and distributed nature of spinal loading and disc deformation, strain patterns were compared qualitatively between groups and across postures to identify consistent trends rather than relying on aggregated scalar summaries.

Statistical analysis

Data analysis was performed using SPSS 21.0 (IBM SPSS Statistics, Armonk, NY, USA). Normality of continuous variables was assessed through skewness/kurtosis indices combined with visual inspection of histograms and Q-Q plots. Disc height parameters are presented as mean ± standard error (SE). Intragroup comparisons (pre- vs. post-fatigue) were analyzed using paired Student’s t-tests, while intergroup comparisons (CLBP vs. HC) employed independent-samples t-tests. A two-tailed α-level of 0.05 defined statistical significance throughout the study.

Results

A total of 24 pain-free adults (males, 9; females, 15; age, 23.54 ± 2.25 years; height:1.70 ± 0.06 m, body mass index (BMI), 21.48 ± 1.42 kg/m²) and 29 patients with CLBP (male, 17; female, 12; age, 24.41 ± 2.44 years; height:1.72 ± 0.08 m, BMI, 21.79 ± 1.92 kg/m²) were included in this study. All participants completed the experimental task. The mean number of lifting cycles completed was significantly lower in the CLBP group (25.31 ± 4.86) than in the HC group (32.08 ± 7.31) (p < 0.001). The participant characteristics and clinical features are shown in Table 1. No significant differences were observed between the CLBP and HC groups in terms of age, height, weight, and BMI. (p > 0.05, Table 1).

Table 1.

Participant characteristics and clinical features

	CLBP	HC	t	P value
Sex (M/F)	17/12	9/15	-	-
Age (years)	24.41 ± 2.44	23.54 ± 2.25	1.34	0.19
Height (m)	1.72 ± 0.08	1.70 ± 0.06	1.27	0.21
Weight (kg)	64.71 ± 9.21	61.98 ± 6.67	1.21	0.23
BMI (kg/m2)	21.79 ± 1.92	21.48 ± 1.42	0.67	0.51
Lifting times	25.31 ± 4.86	32.08 ± 7.31	−4.17	< 0.001
VAS	4.52 ± 1.70	-	-	-
ODI	11.31 ± 11.06	-	-	-

F female, M male, BMI body mass index, VAS visual analogue scale, ODI Oswestry Disability Index

Intervertebral disc morphological characteristics in different postures

Fatigue-induced changes in inter-segmental disc height during posture shifts are summarized in Table 2.

Table 2.

Statistical results of the intervertebral disc height before and after fatigue between people with and without CLBP

Posture	Intervertebral Disc	Intervertebral Disc Height
		CLBP-pre	CLBP-post	HC-pre	HC-post
Neutral	L1-L2	6.25 ± 1.16	6.17 ± 1.09*	6.46 ± 0.81	6.37 ± 1.11
	L2-L3	7.30 ± 0.99	7.34 ± 0.99	7.64 ± 0.95	7.51 ± 0.84
	L3-L4	7.98 ± 1.08	7.79 ± 0.99	8.01 ± 0.98	7.75 ± 0.92
	L4-L5	7.45 ± 0.92	7.57 ± 1.18	8.12 ± 0.93#	7.96 ± 1.00
	L5-S1	5.39 ± 1.53	4.98 ± 1.30**	6.03 ± 1.57	6.07 ± 1.68##
Maximum Extension	L1-L2	6.48 ± 1.02	6.23 ± 1.10*	6.59 ± 1.15	6.63 ± 0.95
	L2-L3	7.32 ± 1.15	7.44 ± 1.10	7.75 ± 0.97	7.72 ± 1.18
	L3-L4	7.98 ± 1.10	7.84 ± 0.96	7.89 ± 0.90	7.80 ± 0.94
	L4-L5	7.69 ± 1.12	7.61 ± 1.15	8.34 ± 1.07#	8.14 ± 1.06
	L5-S1	5.56 ± 1.53	5.47 ± 1.51**	6.20 ± 1.45	6.09 ± 1.78##
~ 45° Flexion	L1-L2	6.45 ± 2.59	5.62 ± 1.62*	6.31 ± 1.54	6.78 ± 1.47
	L2-L3	7.74 ± 1.97	7.83 ± 1.88	7.87 ± 1.29	6.89 ± 1.34
	L3-L4	7.45 ± 1.54	8.08 ± 2.10	7.63 ± 1.86	7.71 ± 2.02
	L4-L5	7.03 ± 2.68	6.99 ± 2.43	7.61 ± 1.95#	8.07 ± 2.13
	L5-S1	5.25 ± 2.35	4.66 ± 2.21**	5.73 ± 2.12	5.85 ± 2.58##

*CLBP-post vs CLBP-pre, p < 0.05

**CLBP-post vs CLBP -pre, p < 0.01

^#CLBP-pre vs. HC-pre, p < 0.05

 ^##CLBP-post vs. HC- post, p < 0.01

L1/L2 segment

CLBP group

Fatigue loading induced significant decreases in L1/L2 disc height (all p < 0.05) under all tested postures: Neutral: 6.25 ± 1.16 mm (pre-fatigue) vs. 6.17 ± 1.09 mm (post-fatigue). ~45° flexion: 6.45 ± 2.59 mm (pre-fatigue) vs. 5.62 ± 1.62 mm (post-fatigue). Maximal extension: 6.48 ± 1.02 mm (pre-fatigue) vs. 6.23 ± 1.10 mm (post-fatigue).

HC group

L1/L2 dimensions remained stable post-fatigue (p > 0.05), with values persistently lower than adjacent caudal segments (L2/3–L4/5).

L2/L3 and L3/L4 segments

Both groups demonstrated non-significant fluctuations in disc height at these levels following fatigue intervention (all p > 0.05).

L4/L5 segment

HC group

Pre-fatigue L4/L5 disc height exceeded CLBP group measurements in all postural configurations (all p < 0.05), potentially indicating early degenerative susceptibility in the CLBP population.

L5/S1 segment

CLBP group

A statistically significant reduction in L5/S1 disc height occurred post-fatigue across all postures (all p < 0.01). Specific changes included: Neutral: 5.39 ± 1.53 mm (pre-fatigue) vs. 4.98 ± 1.30 mm (post-fatigue). ~45° flexion: 5.25 ± 2.35 mm (pre-fatigue) vs. 4.66 ± 2.21 mm (post-fatigue). Maximal extension: 5.56 ± 1.53 mm (pre-fatigue) vs. 5.47 ± 1.51 mm (post-fatigue).

HC group

No statistically significant differences in L5/S1 disc height were observed between pre- and post-fatigue (p > 0.05). Notably, this group maintained consistently greater disc height versus the CLBP cohort, with intergroup differences reaching significance post-fatigue (p < 0.01).

Intervertebral disc strain distribution patterns

Qualitative assessment of IVD strain distribution patterns, particularly shear strain, indicated the common occurrence of asymmetrical deformation along the mediolateral (left-right) axis, although this feature was not formally quantified. Furthermore, disc strain distribution demonstrated distinct segment- and direction-dependent characteristics across different postures, as detailed below:

1. Disc deformation during ~45° flexion demonstrated biomechanically graded responses versus supine positioning (Fig. 3), featuring

   1. Tensile and compressive strain distribution

      • ~45° Flexion Posture: Posterior regions of discs from L1/2 to L5/S1 displayed tensile strain (red), while anterior regions were dominated by compressive strain (blue). Segment-specific differences were observed: anterior compressive strain magnitudes at L4/5 and L5/S1 significantly exceeded those of other segments.
   2. Shear strain characteristics

      • Anterior-directed shear forces were predominant. Anterior regions of L1/2 to L4/5 discs exhibited superimposed forward shear strain on compressive strain. Despite bearing compressive strain spanning ~50% of disc thickness (significantly higher than other segments), the L5/S1 disc demonstrated minimal shear strain (shorter vector arrows), a phenomenon more pronounced in the CLBP cohort.
2. Based on the supine position, the neutral position lumbar disc deformation patterns and mechanical load distribution exhibited significant segment-dependent characteristics (Fig. 4):

   1.  CLBP group

      • Overall pattern: Coexistence of anterior tensile strain and posterior compressive strain with posterolateral shear vectors.
      • L4/5 and L5/S1 segments: Compressive strain magnitudes increased significantly (20–50% of disc thickness), exacerbated post-fatigue. Shear strain amplitudes were markedly reduced compared to the asymptomatic group (shorter vector arrows).
   2. HC group

      • Upper Lumbar Segments (L1/2–L3/4): Anterior disc regions showed mild tensile strain (yellow), while posterior regions demonstrated predominant compressive strain (blue). Shear force vectors concentrated posterolaterally. L4/5 Segment: No significant compressive or shear strain detected (green). L5/S1 Segment: Compressive strain occupied 20–30% of disc thickness, with leftward shear strain direction.
3. Disc deformation during maximum extension, segment-dependent disc deformation and load distribution were observed versus supine positioning: (Fig. 5):

   1. CLBP group

      • Upper Lumbar Segments (L1/2–L3/4): Deformation pattern consistent with HC group (anterior tension/posterior compression), posterolateral shear vectors. L4/5 segment: No significant compressive/tensile strain pre- or post-fatigue (green), with leftward shear strain direction. L5/S1 segment: Compressive strain increased significantly (20–30% of disc height), while shear strain amplitudes were reduced compared to the asymptomatic group (shorter vector arrows).
   2. HC Group

      • Upper Lumbar Segments (L1/2–L3/4): Anterior tensile strain (red) and posterior compressive strain (blue) aligned with stress transmission pathways during lumbar extension, reflecting adaptive responses of the vertebral-facet joint complex to extension moments. Shear vectors were directed posterolaterally. L4/5 segment: Pre-fatigue: Mild anterior tensile strain (yellow) with posterior compression (blue). Post-fatigue: No detectable anterior strain (green). L5/S1 segment: Compressive strain spanned 20–40% of disc thickness, with anterolateral shear strain direction.

Fig. 3.

Intervertebral disc (L1–S1) strain distribution in HC and CLBP groups during ~45° flexion, before and after fatigue. Disc geometries were standardized, and strain values were normalized to disc thickness in the supine position. Color maps indicate tensile strain (red, up to +50%) and compressive strain (blue, down to –50%). Shear strain magnitude and direction are shown by vector arrows (length proportional to magnitude)

Fig. 4.

Strain distribution across L1–S1 discs in HC and CLBP groups in the neutral standing position, before and after fatigue. Standardized disc dimensions and normalization to supine disc thickness were applied. Tensile (+50%, red) and compressive (–50%, blue) strains are represented by color gradients. Shear strain is indicated by vectors (arrow length reflects magnitude)

Fig. 5.

Disc strain distribution (L1–S1) during maximum extension in normal and CLBP groups, pre- and post-fatigue. All discs were geometrically standardized and strain was normalized to supine disc thickness. Color scales denote tensile (red, +50%) to compressive (blue, –50%) strain. Shear strain is depicted using vectors scaled by magnitude

Discussion

This study systematically investigated segment-specific biomechanical mechanisms underlying intervertebral disc degeneration in CLBP patients by analyzing morphological and biomechanical responses across postures and fatigue conditions. The principal findings reveal distinct segmental vulnerability gradients. Specifically, the L5/S1 segment demonstrated significant post-fatigue height reduction across neutral, flexion (~ 45°), and extension postures in CLBP patients, consistently lower than upper lumbar segments (L1-L4). This pronounced vulnerability at the lumbosacral junction likely stems from its role in sustaining greater cumulative vertical compression and shear stresses within the spine-pelvis kinematic chain, coupled with viscoelastic buffering dysfunction in degenerated discs leading to direct stress transmission to the annulus fibrosus [38–40]. Furthermore, the significant fatigue-induced disc height reduction observed at L5/S1 in the CLBP group may be attributed to pre-existing disc degeneration. Although the present study did not include MRI-based degeneration grading, the significantly lower baseline disc height in the CLBP group provides indirect evidence of degenerative changes [1, 2]. Degenerated discs, with compromised viscoelastic properties, are likely more vulnerable to height loss under prolonged cyclic loading. This represents a limitation of the current study, and future research incorporating advanced imaging is warranted to directly correlate degeneration grade with disc mechanical behavior under load. Conversely, the L4/L5 segment exhibited early degeneration sensitivity, evidenced by significantly lower pre-fatigue disc height in CLBP patients compared to controls, potentially reflecting compromised compensatory mechanisms involving muscular co-contraction and ligamentous tension. In contrast, upper lumbar segments (L1-L3) displayed superior mechanical stability and degeneration resistance, attributed to anatomical protection (rib cage, sagittal-oriented facet joints [41]) and favorable load distribution, despite some post-fatigue height reduction at L1/L2.

Distinct segment-specific strain distribution patterns were observed, particularly under flexion. Flexion induced characteristic compression-tension coupling aligning with the posterior shift of the instantaneous axis of rotation, predisposing posterior annular pathology. Crucially, lower lumbar segments (L4-S1) exhibited elevated compressive strain in CLBP patients. This, combined with reduced shear strain dissipation capacity, suggests disordered collagen networks concentrate compressive loads at nucleus-endplate interfaces [42]. Postural responses further differentiated groups: in the upright position, normal controls maintained minimal L4/5 strain indicative of effective muscular-ligamentous stabilization, contrasting with significant compressive strain in CLBP patients reflecting decompensation. Following fatigue, normal controls demonstrated adaptive modification of L4/L5 strain patterns (e.g., anterior strain elimination), while the CLBP group exhibited persistent biomechanical rigidity. A consistent “high-compression/low-shear” phenotype was identified at L5/S1 across postures in CLBP, exacerbated by fatigue and correlating with anatomical factors (sacral tilt, sagittal facet orientation) and impaired proteoglycan-mediated shear dissipation [43].

Previous research indicates that disc microinstability (termed “active discopathy”) by Nguyen et al. [44] is considered the initial stage of the spinal degenerative cascade [45] and implicated in degenerative conditions such as LBP [46]. However, diagnosing this condition remains challenging, and its association with LBP is contentious. Consequently, there is a critical need to develop novel diagnostic approaches to assess a disc’s capacity to withstand compressive loads and its microstability or microinstability.

Although in vivo measurement of intradiscal deformation offers valuable insights into disc pathology and the site-specificity of disc tears and herniations [47], it presents significant technical hurdles, resulting in few reported methods suitable for human application. Byrne et al. proposed a pointwise mapping-based method for quantifying disc biomechanical properties [48]. However, this approach relies on repeated ionizing radiation imaging, rendering it unsuitable for longitudinal studies. Menon et al. introduced a radiation-free method utilizing MRI combined with optical flow analysis [49]. Leveraging MRI’s inherent high soft-tissue contrast, this technique shows greater promise for quantifying intradiscal deformation. Nevertheless, the Menon method employs non-standard MRI sequences and lacks commercial software validation, limiting its clinical utility, and the optical flow algorithm may introduce measurement errors.

Similar to the present study, Menon et al. observed segmental differences in load-induced deformation (greatest compression in the lowest lumbar segments) [49] and noted a highly complex spatial distribution of deformation values across the disc, underscoring the necessity for pixel-level mapping capability. While global measurements provide important data, characterizing deformation within specific internal disc regions may reveal functional abnormalities earlier in the degenerative cascade. Investigating deformation patterns could elucidate why patients with specific spinal morphotypes (e.g., Roussouly Type 1 [50]) are more susceptible to disc degeneration and injury [51].

Although a gold standard for in vivo intradiscal deformation measurement remains elusive, our findings provide indirect support for previous studies [52, 53]. Specifically, the segments exhibiting the greatest compressive deformation (L5/S1) were also those with significantly reduced baseline disc height—a well-established morphological indicator of degeneration. This observation suggests that degenerated discs, characterized by altered matrix composition and impaired fluid exchange, possess diminished viscoelastic properties and undergo greater compressive deformation under sustained loading. Thus, our results align with the existing understanding that more degenerated discs are prone to increased compressive deformation [53]. Known age-related progression of degeneration progressively alters disc composition and structure (e.g., proteoglycan loss reducing osmotic pressure), potentially explaining the decreased compressive stiffness observed in highly degenerated discs [54]. These findings support the biomechanical “vicious cycle” hypothesis of CLBP progression [55]: viscoelastic loss in the lower lumbar segments (L4-S1) triggers compensatory load redistribution to adjacent segments (e.g., L3/L4), establishing a degenerative cycle of “degeneration-compensation-further degeneration.” We also observed pathological decoupling of shear and compressive strains – reduced shear dissipation capacity in degenerated discs concentrates compressive stress at critical endplate-nucleus interfaces [56]. This localized stress concentration provides a plausible biomechanical mechanism for structural damage (e.g., endplate microfractures, subchondral sclerosis), potentially correlating with established imaging biomarkers of degenerative progression.

An interesting qualitative observation was the frequent presence of asymmetric disc deformation, particularly in the mediolateral direction. This pattern is consistent with ex vivo observations by Chan et al. and the in vivo findings of Byrne et al. [48, 52], who postulated that inherent variations in disc geometry (e.g., endplate topography and nucleus positioning) could lead to eccentric load distribution [48]. In vivo, this inherent asymmetry may be exacerbated during dynamic weight-bearing tasks by factors such as asymmetric paraspinal muscle activation, limb length discrepancy, or subtle spinal misalignments. While the present study was not designed to quantitatively analyze the source or prevalence of this asymmetry, its consistent visual presence suggests that symmetric models of disc loading may not fully capture in vivo biomechanics. We hypothesize that subject-specific morphology and loading patterns significantly influence deformation asymmetry. Future research incorporating quantitative asymmetry metrics, coupled with detailed morphological and muscular assessments, is needed to elucidate the clinical significance of this asymmetric deformation behavior and its potential role in the degenerative process.

This study has several limitations that should be considered. First, the cohort included both patients with CLBP and asymptomatic controls, but the limited sample size did not allow for subgroup analyses (e.g., by age, sex, or degeneration severity), which may affect the generalizability of the results. Second, the assessment of fatigue relied on a subjective self-report measure, which introduces inherent subjectivity and potential measurement bias. Optimizing the fatigue induction protocol—particularly by developing quantifiable, individualized loading relative to muscle endurance—would strengthen future investigations. Third, although biplanar fluoroscopy provided high-accuracy kinematic data, it captures only a single biomechanical aspect of complex spinal behavior during functional activities. Therefore, findings related to spinal overload should be interpreted as reflecting potential functional adaptations rather than conclusive biomechanical changes. Finally, the absence of quantitative disc degeneration grading (e.g., Pfirrmann classification via MRI) for both groups must be acknowledged. Although the primary outcome focused on dynamic disc height change under loading—which may be less influenced by baseline degeneration than absolute height—we cannot exclude the possibility that differences in degenerative status contributed to the observed effects. Future research should integrate MRI-based degeneration grading to better control for this confounder and further elucidate how degeneration severity modulates disc biomechanical behavior under fatiguing conditions.

Conclusions and future perspectives

This study quantified segment-dependent biomechanical alterations in individuals with chronic low back pain (CLBP). The findings suggest that the lumbosacral junction (L4–S1) may serve as a region susceptible to degeneration, likely due to its predominance in compressive loading and reduced capacity for shear dissipation. In contrast, the upper lumbar segments appear to maintain relative stability through compensatory anatomical and functional adaptations. These insights offer potential biomechanical support for refining diagnostic approaches—such as focused assessment of L5/S1 dynamics—and for developing targeted rehabilitation strategies, including dynamic stabilization training. Further interdisciplinary studies are needed to clarify the underlying mechano-biological interactions and to explore personalized preventive and management strategies for CLBP.

Abbreviations

CLBP

Chronic low back pain

IVD

Intervertebral disc

DFIS

Dual fluoroscopic imaging system

HC

Healthy control

ICP

Iterative closest point

BMI

Body mass index

VAS

Visual analogue scale

ODI

Oswestry Disability Index

Authors’ contributions

XX designed the research methods, performed the experiments, analyzed the data, and drafted the manuscript. LJ acquired and interpretated the data and critically reviewed the manuscript. WY contributed to critical review of the manuscript. HY, YQ participants recruitment and acquisition of data. LC and YY conceptualized the study, interpreted the data, and revised the manuscript. TYT revised the manuscript. All named authors meet the International Committee of Medical Journal Editors (ICMJE) criteria for authorship, and take responsibility for the integrity of the work as a whole. All authors have read and approved the final submitted manuscript.

Funding

This study and the journal’s Rapid Service Fee were funded by the Fundamental Research Funds for the Central Universities (2023-4-YB-12) and Natural Science Project of Bengbu Medical University (2023byzd199). The funders had no involvement in the study design, data analysis and interpretation of the results. The authors thank all the participants for participating in this study.

Data availability

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

Declarations

Ethics approval and consent to participate

Participation in this study was entirely voluntary. Written informed consent, encompassing both participation and publication of anonymized data, was obtained from all participants prior to study commencement. The study protocol received approval from the Institutional Review Board of Tongji Hospital, Shanghai, China (Protocol No. 2021-011-SK) on January1, 2021. This research was conducted in strict accordance with the ethical principles outlined in the Declaration of Helsinki.

Consent for publication

Written informed consent for publication was obtained from all participants depicted in this study. Identifying details have been anonymized in the images to protect participant privacy.

Competing interests

The authors declare no competing interests.