Skip to main content
NCBI home page
Bone & Joint Research logoLink to Bone & Joint Research
. 2025 Oct 6;14(10):839–849. doi: 10.1302/2046-3758.1410.BJR-2025-0064.R1

Discogenic pain in lumbar vertebra

is quantitative assessment of the osteochondral endplates by simultaneous 18F-NaF PET/MRI helpful?

Yu-Pang Lin 1, Chih-Chien Wang 2, Han-Bin Huang 3, Guo-Shu Huang 1,4, Shih-Wei Chiang 1, Yi-Chih Hsu 1,✉
PMCID: PMC12498161  PMID: 41047132

Abstract

Aims

Discogenic pain due to intervertebral disc (IVD) degeneration contributes substantially towards low back pain (LBP). This study aims to develop novel and non-invasive diagnostic tools by directly comparing the quantitative results of the normal and degenerated osteochondral endplates obtained using simultaneous 18F-NaF PET/MRI.

Methods

We prospectively enrolled 22 participants with chronic lower back pain from August 2022 to November 2023. We classified all 110 lumbar IVDs (L1/2-L5/S1 in 22 people) into 75 normal IVDs and 35 degenerative IVDs. The T2 relaxation times in the cartilaginous endplate on MRI (T2rt-CEP-MR) and normalized subchondral bone on positron emission tomography (PET) images (SUVmax-SB-PET) of all 110 lumbar IVDs were recorded. To analyze the relationship between the cartilaginous endplate and neighbouring subchondral bone, we employed Pearson correlation coefficient and generalized estimating equation (GEE) models to examine the T2rt-CEP-MR and SUVmax-SB-PET results in normal and degenerative IVDs.

Results

We observed lower T2rt-CEP-MR in the middle location of the disc (p = 0.003) and higher SUVmax-SB-PET in the anterior, middle, and posterior locations of the disc (p = 0.030, p = 0.006, and p < 0.001, respectively) in degenerative IVDs. Regarding the association between the T2rt-CEP-MR of discs and the SUVmax-SB-PET of the neighbouring subchondral bone, we observed a significant correlation (r = 0.211 to 0.328, p < 0.001) in all locations of normal IVDs. Additionally, a significant correlation was observed in the anterior and middle locations of the degenerative IVDs, and a marginally significant correlation was observed in the posterior region of the degenerative IVDs (r = 0.159, p = 0.060).

Conclusion

Simultaneous 18F-NaF PET/MRI is a novel and reliable technique for evaluating the osteochondral endplate. The differences in quantitative results could help clinicians to establish an impression of discogenic pain. A spatial relationship exists between the biochemical changes within discs and subchondral bone metabolism.

Cite this article: Bone Joint Res 2025;14(10):839–849.

Keywords: Intervertebral disc, MRI, Positron emission tomography, Spine, Discogenic pain, lumbar vertebrae, intervertebral discs (IVDs), subchondral bone, MR imaging, cartilaginous endplates, PET and MRI, positron emission tomography, lower back pain, clinicians

Article focus

  • To evaluate the reliability of measuring quantitative T2 relaxation times in the cartilaginous endplate on MRI (T2rt-CEP-MR) and standardized uptake value in the subchondral bone on positron emission tomography (PET) images (SUVmax-SB-PET).

  • To investigate the difference of the quantitative results between the normal and degenerated osteochondral units.

  • To explore the relationships between the biochemical changes in cartilaginous endplate and the metabolism in neighbouring subchondral bone.

Key messages

  • The most important findings of this study are that there is high reliability with no remarkable variance in the first and second measurements, different quantitative results between normal and degenerative intervertebral disc degenerations (IVDs), and a significant correlation between the T2rt-CEP-MR of discs and the SUVmax-SB-PET of neighbouring subchondral bone.

Strengths and limitations

  • To the authors’ knowledge, this is the first study to compare normal and degenerative IVDs by simultaneous 18F-NaF PET/MRI, which revealed significantly different quantitative results, as demonstrated by lower T2rt-CEP-MR and higher SUVmax-SB-PET in the degenerative discs, accompanied by a spatial relationship between the biochemical changes within discs and metabolism in subchondral bone.

  • The study is subject to several limitations, including a restricted sample size, the absence of histological correlation, and the lack of interobserver analysis.

Introduction

Persistent lower back pain and its treatment is a considerable public health issue, marked by high economic costs and considerable psychosocial consequences.1 The primary factor contributing to chronic back pain is the degeneration of the intervertebral discs (IVDs).2 IVD degeneration is characterized by the structural deterioration of the disc and reduction in proteoglycan and water levels within the extracellular matrix.3,4 Facet tropism may play a role in the development of disc herniation and alterations in intradiscal pressure, which could further aggravate disc degeneration as a result of modified force distribution and heightened mechanical stress.5 Transcription factors, microRNAs, and potential pharmacological agents that modulate extracellular proteins play a crucial role in the pathogenesis of IVD degeneration.6 The degeneration of IVDs has been the subject of long-standing speculation, with insufficient nutrient delivery to the discs being proposed as a potential contributing factor to its pathogenesis,7,8 associated with abnormal spinopelvic motion.9 A recent study proposed that monomeric isoform CRP is present in IVD tissues and persistently induces proinflammatory and catabolic factors in annulus fibrosus (AF) and nucleus pulposus (NP) cells.10

IVDs lack blood vessels. Therefore, for the transfer of oxygen and nutrients, the cells within the NP primarily depend on the capillary network located in the nearby osteochondral endplate, which consists of subchondral bone (SB) and cartilaginous endplate (CEP).11,12 Defects in the SB-CEP can result in desiccation, depletion of matrix proteins, and structural breakdown of the IVDs.13,14 The interconnection of capillaries between the SB and CEP, acting as a collaborative osteochondral entity, is crucial in preserving adequate disc hydration. A previous study, using dynamic contrast-enhanced MRI, revealed significant changes in the perfusion and diffusion of the SB-CEP with increasing disc degeneration.15 However, the metabolic activity of the SB and the biochemical changes in CEP during disc degeneration have not been fully investigated.

The use of simultaneous positron emission tomography-MRI (PET-MRI) presents a valuable method for evaluating this mechanism in vivo. This approach enables a comprehensive evaluation of CEP through iterative decomposition of water and fat with echo asymmetry and least-squares estimation (IDEAL) and fat-saturated spoiled gradient echo (SPGR) MRI. It also allows the assessment of the SB metabolism through the examination of 18F-sodium fluoride (18F-NaF) uptake using PET.16,17 A previous investigation used SPGR imaging to examine the lumbar CEPs of cadavers, identifying morphological irregularities in the CEP of all lumbar spine specimens and validating it through anatomical inspection.18 T2 mapping demonstrated sensitivity to alterations in collagen and water content. This serves as a dependable and accurate technique for quantitatively assessing the changes in biochemical composition within the CEP and IVD of the spine.16,1918F-NaF accumulates in areas of actively mineralizing bone by interacting with the exposed hydroxyapatite through ion exchange,20 and its presence is associated with histomorphometric indicators of bone turnover, such as mineral apposition rate.21 The 18F-NaF PET imaging technique has been employed for evaluating the metabolic activity of the vertebral endplate.17,22

A previous study proposed a notable association between heightened SB metabolism, as identified through 18F-NaF PET/MRI, and damage to nearby cartilage, as revealed by T2 mapping MRI in individuals diagnosed with osteoarthritis related to anterior cruciate ligament injury.23 This finding demonstrated that changes in SB metabolism may be indicative of structural alterations in the adjacent cartilage. A recent investigation presents a novel therapeutic strategy that integrates vertebral intraosseous and intradiscal injections of plasma rich in growth factors (PRGF-Endoret) for the management of intervertebral disc degeneration and subchondral bone damage. The findings indicate a partial regression of protruded discs and a notable reduction in intravertebral herniations.24 Therefore, this study aimed to assess the feasibility and reliability of a technique for quantitatively assessing T2 relaxation times in the cartilaginous endplate on MRI (T2rt-CEP-MR) and standardized uptake value in the subchondral bone on PET images (SUVmax-SB-PET). We explored the differences in the quantitative results between the normal and degenerated osteochondral endplates. Furthermore, we investigated the relationship between the biochemical changes in the CEP and the metabolism in neighbouring SB.

Methods

Study population

This study was approved by the institutional review board of our organization and written informed consent was obtained from all participants. The research cohort comprised 22 individuals, including 17 males and five females, who reported experiencing symptoms of low back pain within the preceding year, exhibited negative findings on lumbar spine radiographs, and had not undergone any medical treatment for the spinal condition. Individuals who were pregnant, or had a history of diabetes, smoking, cancer, spondylolisthesis, scoliosis, previous lumbar surgery, compression fractures, or who were currently taking osteoporosis medication, were excluded from this study. The age and BMI of all participants were documented. The mean age at the time of undergoing MRI was 38.8 years (SD 11.9). 18F-NaF PET/MRI was performed to analyze the osteochondral endplates of the lumbar vertebrae.

Simultaneous 18F-NaF PET/MRI protocols

All participants underwent 18F-NaF PET/MRI examination using the standardized imaging protocol.25 The PET and MRI data were acquired simultaneously using a hybrid 3.0 T PET-MR imaging unit (GE Healthcare, USA) for 60 minutes.

PET imaging was performed on the lumbar spine, spanning from L1 to S2 vertebrae, using two PET beds with a field of view (FOV) of 26 cm. Approximately 90 MBq 18F-NaF was intravenously injected, and imaging was performed for approximately 50 minutes post-injection.

A simultaneous MRI of the lumbar spine was performed alongside PET acquisition, using two flexible 16-channel receive-only coils (Neocoil, USA). A two-point Dixon acquisition method was employed for MR attenuation correction (MRAC) of the PET data. The morphological and structural characteristics of the spine were evaluated using MRI techniques (Figure 1).

Fig. 1.

Four sagittal MRI scans of the spine labeled IDEAL-SPGR, T2 mapping, and T2WI, showing vertebrae and intervertebral discs. Four sagittal MRI scans of the spine. Two images labeled IDEAL-SPGR show detailed views of vertebrae and intervertebral discs. One image labeled T2 mapping includes a scale bar indicating varying values across the spine. The final image, labeled T2WI, highlights specific vertebrae and discs using arrowheads. These scans illustrate structural and compositional differences in spinal tissues across imaging techniques.

Open in a new tab

The cartilaginous endplates (CEP) and intervertebral discs (IVDs) of a 20-year-old female were examined in the study. a) The CEP exhibits high signal intensity, as indicated by arrowheads, on the median sagittal image obtained using iterative decomposition of water and fat with echo asymmetry and least squares estimation (IDEAL)-fat-saturated spoiled gradient echo (SPGR), with b) a region of interest highlighted on the IDEAL-SPGR image. c) The corresponding CEP area is identifiable on the T2 mapping pseudo-colour map based on the IDEAL-SPGR image. d) Additionally, the median sagittal image of T2WI reveals Pfirrmann grade I in the L3–S1 IVDs, with low signal intensity observed on the CEP (arrowheads).

All MRI data were acquired using a cervical-thoracic-lumbar (CTL) spine coil. Participants were positioned supine on the CTL spine coil, with the alignment line adjusted to correspond to the third lumbar vertebra. The scanning parameters used for the T2-weighted imaging (T2WI) sequence, specifically the Fast Recovery Fast Spin Echo (FRFSE) sequence, included a time of repetition (TR) of 4,000 ms, time of echo (TE) of 121 ms, two excitations (NEX), a slice thickness of 4 mm with a slice gap of 0.5 mm, ten slices, a field of view (FOV) of 220 × 220 mm, a matrix size of 352 × 320, and a scanning time (ST) of two minutes and 16 seconds. The T1WI sequence, specifically the Fluid Attenuated Inversion Recovery (FLAIR) sequence, employed specific imaging parameters including a TR of 2,950 ms, TE of 22 ms, inversion time (TI) of 1,050 ms, a flip angle of 142°, FOV of 220 × 220 mm2, voxel matrix size of 320 × 224, slice thickness of 1.5 mm with a 0.5 mm slice gap, ten slices acquired, two NEX, and a ST of two minutes and 55 seconds. To examine the lumbar CEP in vivo, the IDEAL-SPGR sequence was used as it is capable of providing precise visualization of the structure and positioning of the osteochondral endplates, as depicted in Figure 1a.16 The imaging parameters for IDEAL-SPGR consisted of a TR of 6.7 ms, a TE of 3.0 ms, a matrix size of 320 × 224, a FOV of 220 × 220 mm2, a slice thickness of 3 mm with no slice gap, 12 slices acquired, three NEX, a flip angle of 15°, and a ST of four minutes and 30 seconds. To understand the structural properties of the lumbar IVDs, the L1-2, L2-3, L3-4, L4-5, and L5-S1 discs were selected as regions of interest (ROI), and T2 quantification sequences of MRI in the sagittal plane were acquired.16 The T2 mapping technique employed a multi-echo SE sequence with eight echo acquisitions. The sequence parameters included a TR of 1,000 ms, eight different TE values (8.5, 16.9, 25.4, 33.9, 42.3, 50.8, 59.2, and 67.7 ms), a slice thickness of 3 mm with a 1 mm slice gap, acquisition of ten slices, two NEX, a FOV of 220 × 220 mm², a matrix size of 256 × 192, and an ST of six minutes and 26 seconds.16

Image acquisition

18F-NaF images were generated by reconstructing data obtained from 30 minutes of list mode acquisition, using a time-of-flight PET reconstruction method. The reconstruction process involved four iterations and 28 subsets, with a FOV of 50 cm, a matrix size of 192 × 192, and a slice thickness of 2.78 mm. Various corrections were applied during reconstruction, including scatter, random counts, dead time, and point-spread function. MRAC images were used for automated anatomical segmentation, with tissue-specific linear attenuation coefficients assigned to each segmented tissue.

In the post-processing workstation, the T2 mapping module was chosen within the FuncTool software (GE Healthcare, USA), with a confidence interval established at 95% on both ends and a colour threshold of 20 ms. IDEAL-SPGR images offer a clear depiction of CEP profiles, as referenced in previous literature.16 Subsequently, a ROI was delineated on the IDEAL-SPGR images, as illustrated in Figure 1b. This ROI was then aligned with the corresponding T2 mapping slice to determine the T2 values, as depicted in Figure 1c.

Classification and measurement of the image data

Two radiologists (YCH, GSH) with varying levels of expertise independently analyzed the images, using the double-blind technique. Images displaying lumbar Modic type endplate changes or Schmorl nodes on T1WI and T2WI were not included in the analysis. The Pfirrmann (Pm) grading system,26 which consists of a five-point scale, was established by the two radiologists. The original images were analyzed independently by reader 1 and reader 2, and if their interpretations differed, they collaborated to reach an agreement for the final grading. This evaluation was conducted based on median sagittal images from T2-weighted imaging, to categorize and assess the IVD. Grade I IVD according to the Pm grading system is characterized by a normal appearance with a uniform structure, a bright hyperintense white signal intensity, distinct demarcation between the NP and the annulus fibrosus (AF), and a normal disc height (Figure 1d). Pm grade II–V IVDs, characterized by a non-homogeneous structure, darker signal intensity, blurred boundaries between the NP and the AF, and decreased height, were defined as degenerative IVDs. The IDEAL-SPGR images effectively depict the contour of the CEP, facilitating the delineation of the ROI on these images (Figure 2). Due to the observed decrease in thickness of the CEP and SB in adults, the ROI was restricted to 4 mm2, with careful consideration to avoid the vertebral cortex or disc region. Following this guideline, measurements were conducted twice for each CEP and SB, and the mean value was calculated. Prior to the execution of the measurements, reader 3 underwent comprehensive training, which encompassed the careful selection of locations and the methodologies employed in the selection process, thereby ensuring the reliability of the results obtained. The third radiologist (reader 3) independently drew the ROIs of the targets within the CEP, SB, and the CSF region at the L5 vertebral level, without prior knowledge of the specific group to which the patients belonged. The SUVmax of the CSF region (SUVCSF) at the L5 vertebral level was used as an internal control, and the SUVmax was then normalized by dividing it with the SUVCSF at L5.27 Thereafter, the third radiologist (YPL) recorded T2 relaxation times in CEP on MRI (T2rt-CEP-MR) and normalized the maximum standardized uptake value in SB on PET images (SUVmax-SB-PET). The measurements were repeated one month later to evaluate intraobserver variations. The study design is depicted in Figure 3, providing a visual aid for a thorough examination of the workflow and methodologies employed in this study.

Fig. 2.

A sagittal spine diagram with four labeled columns. The final column includes T2 mapping and PET imaging with corresponding MRI scans. The figure presents a sagittal view of the spine divided into four vertical columns labeled A through D. Columns A, B, and C contain schematic diagrams of vertebrae with dashed lines marking specific measurement points labeled a through d. Column D displays advanced imaging results, including T2 mapping of the cartilaginous endplate and PET imaging of the subchondral bone. To the right of these mappings, two MRI scans provide detailed anatomical views that correspond to the imaging data shown in column D.

Open in a new tab

Illustration depicting the region of interest (ROI) configuration of the cartilage endplates and subchondral bone within the osteochondral endplate. The ROI was aligned with the corresponding T2 mapping slice for the determination of T2 values. a) Initially, a line denoted as “a” was drawn on the maximum amplification of the IDEAL-spoiled gradient echo (SPGR) median sagittal image, by connecting the midpoint of the anterior and posterior margins of each vertebral body. Subsequently, a perpendicular line labelled “b” was established to intersect with the upper and lower cartilaginous endplate (CEP), at a single point individually. b) The lines “c” and “d”, which were orthogonal to the anterior and posterior edges of vertebrae, respectively, were parallel to line “b”, and also intersected with the upper and lower CEP. c) There were six CEP intersections of lines “b", "c", and "d” on each intervertebral disc. d) The positions of the intersection of these CEP points on MRI and the corresponding subchondral bone below each CEP point on PET images were then determined.

Fig. 3.

A flowchart showing the classification, imaging, and statistical analysis process for intervertebral discs from 22 participants. A flowchart outlining the process used to classify, measure, and analyze intervertebral discs from 22 participants. It begins with the identification of lumbar and sacral vertebrae and their corresponding discs. Structure classification is performed by three readers using Pfirrmann grading, with degenerative discs identified separately. Imaging includes T2 mapping of the cartilaginous endplate and PET scans of subchondral bone, along with location recognition, region-of-interest selection, and specific imaging metrics. Statistical analysis divides the discs into normal and degenerative groups, each assessed for metabolic and biomechanical activity of the adjacent osteochondral endplate.

Open in a new tab

The process entailed in the workflow of formulating this research. Overall, 22 participants displayed no indications of lumbar ailments aside from degenerative disc disease. Ten cartilaginous endplates (CEPs) for each participant were selected from the lower level of the CEP and subchondral bone (SB) of L1 to the upper level of the CEP of S1. This research investigation was carried out involving a total of 110 intervertebral discs (IVDs), consisting of 75 IVDs classified as normal and 35 IVDs classified as degenerative. These IVDs were evaluated through subsequent imaging measurements and statistical analysis. PET, positron emission tomography; ROI, region of interest; SB, subchondral bone.

Statistical analysis

We performed inferential statistical analyses using the chi-squared test, paired t-test, Bonferroni test, and generalized estimating equation (GEE) model.

To assess the differences between the first and second measurements for T2rt-CEP-MR and SUVmax-SB-PET on lumbar osteochondral units, we used the paired t-test.

We determined the mean and SD of biochemical changes in T2rt-CEP-MR and bone metabolism in SUVmax-SB-PET for the normal and degenerative IVD groups. Additionally, we employed the chi-squared test to explore the linear trends across different levels of osteochondral endplates of the lumbar vertebrae. We tested the hypothesis that T2rt-CEP-MR and SUVmax-SB-PET differed between different locations of the osteochondral endplates concerning the lumbar spine using the Bonferroni test.

The relationship between T2rt-CEP-MR and SUVmax-SB-PET in the osteochondral endplates of the normal IVDs and degenerative IVDs in each subject for each compartment was analyzed using the Pearson correlation coefficient. We also compared the biochemical changes in the CEP and the metabolic activity of the SB of the osteochondral endplates between the normal and degenerative IVD groups using the GEE models after adjusting for age, sex, BMI, and the location of the endplate. Statistical analyses were performed using SPSS 20.0 (IBM, USA). Statistical significance was set as a p-value < 0.05.

Results

Simultaneous 18F-NaF PET/MRI is technically successful for analyzing the osteochondral endplates of all lumbar IVDs. The lumbar IVDs were categorized into two groups: 75 normal IVDs and 35 degenerative IVDs. In both groups, no remarkable variance was detected in T2rt-CEP-MR and SUVmax-SB-PET between the first and second measurements at the anterior, middle, and posterior locations of osteochondral endplates (Table I). The inter-level differences (from L1/2 to L5/S1) of both T2rt-CEP-MR and SUVmax-SB-PET in osteochondral endplates were not statistically significant (Figure 4; p = 0.452, p = 0.558, p = 0.639, and p = 0.292, chi-squared test).

Table I.

Intrarater reliability of the measurements of biochemical changes of the cartilage endplate and metabolic activity of the subchondral bone (SB) on lumbar osteochondral endplates.

Group Location of endplate Anterior Middle Posterior
Measurement Mean (SD) p-value Mean (SD) p-value Mean (SD) p-value
First measure Second measure First measure Second measure First measure Second measure
Normal IVD T2rt-CEP-MR 47.32 (10.21) 47.32 (10.17) 0.998 63.39 (11.70) 63.40 (11.67) 0.990 49.68 (10.58) 49.63 (10.62) 0.962
SUVmax-SB-PET 2.622 (1.650) 2.627 (1.647) 0.984 3.400 (1.902) 3.408 (1.904) 0.951 2.576 (1.636) 2.566 (1.367) 0.962
Degenerative IVD T2rt-CEP-MR 48.96 (11.57) 48.98 (11.44) 0.976 58.30 (10.94) 58.36 (10.89) 0.968 51.42 (11.86) 51.38 (11.80) 0.989
SUVmax-SB-PET 3.163 (1.795) 3.152 (1.789) 0.977 3.951 (1.817) 3.964 (1.827) 0.969 3.246 (1.545) 3.249 (1.534) 0.972
Open in a new tab

All p-values were calculated using paired t-test.

CEP, cartilaginous endplate; IVD, intervertebral disc; SUVmax-SB-PET, normalized maximum standardized uptake value in subchondral bone on positron emission tomography; T2rt-CEP-MR, T2 relaxation times in cartilage endplate on MR.

Fig. 4.

Four box plots compare imaging values for normal and degenerative intervertebral discs across multiple spinal locations. Four box plots arranged in two rows and two columns. The top row shows T2rt-CEP-MR values, and the bottom row shows SUVmax-SB-PET values. The left column represents data from normal intervertebral discs, while the right column represents degenerative discs. Each plot includes data from specific osteochondral endplate locations labeled L12 through S11. The box plots display median values, interquartile ranges, whiskers extending to 1.5 times the IQR, and individual outliers. Trend p-values are noted above each plot, indicating no statistically significant trends across locations for either imaging metric in both disc types.

Open in a new tab

Graphical representation of data using a box plot. The linear trends of T2 relaxation times in cartilage end plates on MRI (T2rt-CEP-MR) and normalized maximum standardized uptake value in subchondral bone on positron emission tomography (PET) images (SUVmax-SB-PET) for the osteochondral endplate of the lumbar vertebrae in the normal and degenerative intervertebral disc (IVD) groups. All p-values calculated using chi-squared test.

Table II displays the mean values and variability represented by the SD of T2rt-CEP-MR and SUVmax-SB-PET for the osteochondral endplates of the lumbar vertebrae in the normal and degenerative IVD groups. Both T2rt-CEP-MR and SUVmax-SB-PET were increased in the middle portion of the osteochondral endplate compared to the anterior and posterior portions of the osteochondral endplate in the normal and degenerative IVD groups (p < 0.001 in T2rt-CEP-MR; p ≤ 0.007 in SUVmax-SB-PET, Bonferroni test; Figure 5). The increase in T2rt-CEP-MR was more pronounced in the posterior locations compared with the anterior locations in the normal (p = 0.010; Figure 5a) and degenerative IVD groups (p = 0.003; Figure 5c). However, no significant difference in SUVmax-SB-PET was observed between the anterior and posterior locations of the osteochondral endplates in the normal (p = 0.739; Figure 5b) and degenerative IVD groups (p = 0.520; Figure 5d).

Table II.

Mean and significance of biochemical changes in the cartilage endplate (CEP) and metabolic activity of the subchondral bone in the normal and degenerative intervertebral disc (IVD) groups.

Group Location of endplate p-value*
Anterior Middle Posterior A vs M P vs M A vs P
Mean (SD) Mean (SD) Mean (SD)
Normal IVD
T2rt-CEP-MR, ms 47.32 (10.17) 63.40 (11.66) 49.65 (10.58) < 0.001 < 0.001 0.010
SUVmax-SB-PET 2.625 (1.646) 3.404 (1.900) 2.571 (1.634) < 0.001 < 0.001 0.739
Degenerative IVD
T2rt-CEP-MR, ms 48.97 (11.46) 58.33 (10.88) 51.40 (11.79) < 0.001 < 0.001 0.003
SUVmax-SB-PET 3.157 (1.786) 3.958 (1.815) 3.247 (1.534) < 0.001 0.007 0.520
Open in a new tab
*

Bonferroni test.

ms, millisecond; SUVmax-SB-PET, normalized maximum standardized uptake value in subchondral bone on positron emission tomography; T2rt-CEP-MR, T2 relaxation times in cartilage endplate on MR.

Fig. 5.

Four box plots compare imaging values for normal and degenerative intervertebral discs at anterior, middle, and posterior positions. Four box plots labeled A through D. Plots A and B show data for normal intervertebral discs, while plots C and D show data for degenerative discs. Plots A and C display T2rt-CE-MR values, and plots B and D show SUVmax-SB-PET values. Each plot compares values at anterior, middle, and posterior positions of the spine. Statistical significance is indicated for each position, with p-values suggesting notable differences in imaging metrics across locations, especially in anterior and posterior regions for both disc types.

Open in a new tab

Box plots showing the comparison of T2 relaxation times in cartilage endplates on MRI (T2rt-CEP-MR) and normalized maximum standardized uptake values in subchondral bone on positron emission tomography (PET) images (SUVmax-SB PET) in the anterior, middle, and posterior portions of the osteochondral endplate within the a) and b) normal group and c) and d) degenerative intervertebral disc (IVD) group.

We compared the biochemical changes in the CEP and metabolic activity within the SB of the osteochondral endplates between the normal and degenerative IVD groups (Table III). After adjusting for age, sex, and endplate location in GEE models, we observed a statistically significant increase in SUVmax-SB-PET values in the anterior, middle, and posterior endplates in the degenerative IVD group compared to the normal IVD group (β = 0.197, β = 0.256, β = 0.328; p = 0.030, p = 0.006, p < 0.001, respectively; GEE model). Additionally, lower T2rt-CEP-MR values on the middle endplate were noted in the degenerative IVD group compared to the normal IVD group (β = -0.078, p = 0.003, GEE model). No significant differences in T2rt-CEP-MR of the anterior and posterior endplates were observed between the degenerative and normal IVD groups (p = 0.739, p = 0.520, GEE model).

Table III.

Comparison of the biochemical changes in the cartilage endplate (CEP) and metabolic activity in the subchondral bone of the osteochondral endplates between the normal and degenerative intervertebral disc (IVD) groups.

Location of endplate Biomechanical activity and metabolism of osteochondral endplate Degenerative IVD (normal IVD as reference) p-value*
β SE
Anterior T2rt-CEP-MR 0.041 0.027 0.127
SUVmax-SB-PET 0.197 0.091 0.030
Middle T2rt-CEP-MR -0.078 0.027 0.003
SUVmax-SB-PET 0.256 0.093 0.006
Posterior T2rt-CEP-MR 0.025 0.029 0.375
SUVmax-SB-PET 0.328 0.096 < 0.001
Open in a new tab

Adjusted for age, sex, BMI, and location of endplate (upper to lower).

*

Generalized estimating equation model.

SE, standard error; SUVmax-SB-PET, normalized maximum standardized uptake value in subchondral bone on positron emission tomography; T2rt-CEP-MR, T2 relaxation times in cartilage endplate on MR.

A spatial relationship was also noted regarding the correlation of the quantitative results between metabolic and biochemical changes in the lumbar osteochondral units. Figure 6 displays a typical instance of elevated SUVmax-SB-PET in close proximity to a region of heightened T2rt-CEP-MR within the cohort of healthy IVDs. In the normal IVD group, a statistically significant correlation was noted between the SUVmax-SB-PET and the adjacent T2rt-CEP-MR across the anterior, middle, and posterior locations (r = 0.211 to 0.328, p < 0.001; Figures 6a to 6c). In the degenerative IVD group, we observed a similar correlation in the anterior (r = 0.231, p = 0.006; Figure 6d) and middle portions (r = 0.238, p = 0.005; Figure 6e) of lumbar osteochondral endplates. We also observed a marginally significant correlation between the SUVmax-SB-PET and the adjacent T2rt-CEP-MR at the posterior portion of the lumbar endplate in the degenerative IVD group (r = 0.159, p = 0.060; Figure 6f).

Fig. 6.

Six scatter plots show the relationship between two imaging values for normal and degenerative intervertebral discs at anterior, middle, and posterior endplate positions. Six scatter plots labeled A through F, each comparing SUVmax-SB-PET and T2rt-CEP-MR values. Plots A, B, and C display data for normal intervertebral discs at anterior, middle, and posterior endplate positions, respectively. Plots D, E, and F show data for degenerative discs at the same positions. Each plot includes a Pearson correlation coefficient and a p-value, indicating the strength and significance of the relationship between the two imaging metrics. Most plots show statistically significant correlations, with the exception of the posterior endplate in degenerative discs.

Open in a new tab

Correlations (r) of T2 relaxation times in cartilage endplates on MRI (T2rt-CEP-MR) and normalized maximum standardized uptake values in subchondral bone, on positron emission tomography (PET) images (SUVmax-SB-PET) of the osteochondral endplate according to the normal intervertebral disc (IVD) and degenerative IVD groups. a) to c) In the normal IVD group consisting of 75 subjects, data points were collected from the anterior, middle, and posterior regions. Each data point represents one of the two osteochondral endplates evaluated, resulting in a total of 150 data points across the normal IVD group. In the group of 35 subjects with degenerative IVDs, data were gathered from the d) anterior, e) middle, and f) posterior regions (70 points total).

Discussion

We showed that simultaneous 18F-NaF PET/MRI can quantitatively measure the metabolic and biochemical activity of the osteochondral endplates in patients. In normal and degenerative IVDs, the biochemical activity of the CEP detected on MRI was spatially correlated with metabolic uptake of the SB on 18F-NaF PET. We observed a reduction in metabolic and biochemical activities at the anterior and posterior regions of the osteochondral endplate of the lumbar vertebrae compared to the middle region in both normal and degenerative lumbar IVDs. Furthermore, the degenerative IVDs, at risk for developing IVD herniation, showed higher SB metabolism at all locations within CEP and lower biochemical activity at the middle location of CEP compared to the normal IVDs. Different measurement results within various locations of the osteochondral endplate may indicate the anatomical modifications in the IVD. Defects in the SB-CEP have been implicated in the structural failure of the IVD.13,14 Nevertheless, the assessment of the metabolic or biochemical bone markers is challenging. The most reliable method for evaluating the synergistic osteochondral unit of the SB-CEP through bone biopsy is expensive, invasive, and constrained in terms of spatial coverage. PET studies using 18F-NaF offer a quantitative assessment of the metabolic activity that exhibits an association with the histomorphometric parameters of bone formation in the typical bone structure.21 MRI can clearly image the CEP on the lumbar osteochondral endplates of cadavers using IDEAL-SPGR sequences,18 and can quantitatively measure the changes in biochemical components of the lumbar CEP using T2 mapping imaging technology.16,19 We found that the intraobserver variability for T2rt-CEP-MR and SUVmax-SB-PET was extremely low, indicating that the analyzed values have high reliability. Altogether, these studies suggest that 18F-NaF PET/MRI can be used to assess the synergistic osteochondral unit of the SB-CEP, and thereby may function as a viable alternative for assessing histomorphometric parameters of the osteochondral unit in the context of invasive bone biopsy procedures.

In the normal and degenerative IVDs, we found that the metabolic changes in the SB and the biochemical activity of the CEP do not change significantly at different lumbar levels. However, both T2rt-CEP-MR and SUVmax-SB-PET exhibited lower values at the anterior and posterior portions of the osteochondral endplates compared to the middle location. This might be because IVDs at different levels have similar anatomical structures, with the NP in the central part and the AF in the peripheral part. The NP has more water and proteoglycan (mean proportion 77% (SD 15%)) and less collagen (4%), while the AF has less water and proteoglycan (mean proportion 70% (SD 5%)) and more collagen (15%).28 The nutrients and oxygen supplied to different components of the IVD also differ, hence the metabolism and biochemistry within the SB-CEP of osteochondral endplate also differ in the central and peripheral parts. An increase in the T2 value on MRI reflects the biochemical activity of the damaged cartilage.29 Moreover, it may also occur early during the development of the disease, and its value does not increase as much as the mechanical degradation of the cartilage.30 Previous studies have demonstrated increased shearing forces over the posterior site of the AF in normal and degenerative IVDs.31,32 This could explain our observation of a more pronounced T2rt-CEP-MR in the posterior portion compared with the anterior portion in the normal and degenerative IVD groups.

IVDs are avascular; therefore, the blood supply pathway of IVDs is crucial. IVDs receive their blood supply predominantly from capillaries that arise in the vertebral bodies, traverse the SB, and end slightly before reaching the CEP. This vascular network serves to deliver essential nutrients and eliminate metabolic byproducts from the IVDs.11,12 Nutrients must then diffuse from the capillaries through the CEP and the dense extracellular matrix of the IVD.28 Hence, the metabolic activity of the SB and the biochemical changes in the CEP should have good synergy. In our study, a correlation was observed between the metabolic changes in the SB and the biochemical activity of the CEP in the normal IVD. However, since IVD degeneration could be due to various factors, such as compromised nutritional supply, mechanical load, and genetic factors, the observed correlation between the metabolic changes in the SB and the biochemical activity of the CEP in degenerative IVDs may not necessarily be a strong one.

In our study, the metabolic activity of the SB of the degenerative IVD was more pronounced at all regions of the osteochondral endplate compared to that of the normal IVD. This suggests that IVD degeneration results in a notable rise in blood circulation within the SB of the osteochondral endplate, thereby enhancing the delivery of nutrients to the IVD. Existing literature indicates that the application of siDDIT4@G5-P-HA hydrogel has demonstrated efficacy in alleviating IVDD in rat models.33 The intraosseous administration of growth factors may serve as a therapeutic approach for degenerative disc disease.24 These observations imply that the integration of PET-MRI could potentially transform the current management strategies for degenerative conditions affecting IVDs and subchondral bone. In cases where a patient presents with low back pain and exhibits abnormalities identified through this imaging modality, clinicians may employ vertebral intraosseous and intradiscal injection techniques as a treatment option. By identifying early metabolic alterations prior to the onset of irreversible structural damage, healthcare professionals may be positioned to initiate preventive or therapeutic measures.

Conversely, the biochemical changes in the CEP of the degenerative IVD were found to be significantly lower at the middle portions of the osteochondral endplates, although no changes were observed in the anterior and posterior portions of the osteochondral endplate compared to the normal IVD. The primary biochemical alteration observed in degenerative IVDs is the reduction in proteoglycan content, which leads to a decline in the osmotic pressure within the disc matrix, resulting in reduced hydration levels.34 A possible explanation is that the NP has more proteoglycan than the AF, hence biochemical changes in the CEP at the middle portion of the osteochondral endplate are more prominent.

One primary constraint of our study was the limited sample size, which led to uneven disc distribution (75 normal IVDs, 35 degenerative IVDs). This limitation also precluded the evaluation of patients exhibiting varying degrees of low back pain. Furthermore, we have not performed studies to correlate PET uptake and T2 relaxation times with the histological alternations in the IVDs of the lumbar vertebrae. Nonetheless, the correlation between different classes of degenerative IVDs, graded using the Pfirrmann MR classification system and histopathological changes, was found to be statistically significant.35 A further limitation of this study is its reliance solely on intraobserver analysis, without the inclusion of interobserver assessment. While intraobserver evaluation can effectively reduce variability among different observers and mitigate subjective bias, making it more appropriate for preliminary validation, the lack of interobserver analysis constrains the generalizability of the findings across various evaluators.36 Future studies can include interobserver comparisons to ensure broader applicability.

The concurrent use of simultaneous 18F-NaF PET/MRI demonstrates technical viability and holds potential as a method for evaluating various metabolic and biochemical parameters of the osteochondral endplates in the lumbar vertebrae, along with the spatial interconnections between these tissues. This method is promising for finding biomarkers that can quantitatively evaluate IVD degeneration, a critical aspect in the development and assessment of disease-modifying therapies aimed at halting or reversing the advancement of the disease. The integration of 18F-NaF PET-MRI into clinical practice could allow for earlier interventions, more accurate disease monitoring, and personalized treatment planning.

Author contributions

Y. Lin: Data curation, Investigation, Writing – original draft

C. Wang: Data curation, Resources

H. Huang: Formal analysis, Methodology

G. Huang: Resources, Supervision

S. Chiang: Data curation, Methodology

Y. Hsu: Conceptualization, Investigation, Methodology, Writing – original draft, Writing – review & editing

Funding statement

The author(s) disclose receipt of the following financial or material support for the research, authorship, and/or publication of this article: the work was supported by grants from the Ministry of Science and Technology of Taiwan (MOST 111-2314-B-016-031). No funding was received from GE Healthcare for publication activities.

ICMJE COI statement

All authors report grants from the Ministry of Science and Technology of Taiwan (MOST 111-2314-B-016-031), related to this study. Each author certifies that there are no funding or commercial associations (consultancies, stock ownership, equity interest, patent/licensing arrangements, etc.) that might pose a conflict of interest in connection with the submitted article related to the author or any immediate family members. No funding was received from GE Healthcare for publication activities.

Data sharing

The datasets generated and analyzed in the current study are not publicly available due to data protection regulations. Access to data is limited to the researchers who have obtained permission for data processing. Further inquiries can be made to the corresponding author.

Acknowledgements

We thank GE Healthcare, Taiwan, for providing sequence information/optimization and technical support for this study.

Ethical review statement

The authors declare that the work described has been performed in accordance with the Declaration of Helsinki of the World Medical Association revised in 2013 for experiments involving humans. The Institutional Review Board for Human Investigation (TSGHIRB B202205009) approved this study protocol. The requirement for written informed consent was waived.

Open access funding

The open access fee for this article was self-funded.

© 2025 Lin et al. This is an open-access article distributed under the terms of the Creative Commons Attribution Non-Commercial No Derivatives (CC BY-NC-ND 4.0) licence, which permits the copying and redistribution of the work only, and provided the original author and source are credited. See https://creativecommons.org/licenses/by-nc-nd/4.0/

Data Availability

The datasets generated and analyzed in the current study are not publicly available due to data protection regulations. Access to data is limited to the researchers who have obtained permission for data processing. Further inquiries can be made to the corresponding author.

References

  • 1. Mathew J, Singh SB, Garis S, Diwan AD. Backing up the stories: the psychological and social costs of chronic low-back pain. Int J Spine Surg. 2013;7:e29–38. doi: 10.1016/j.ijsp.2013.02.001. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2. Luoma K, Riihimäki H, Luukkonen R, Raininko R, Viikari-Juntura E, Lamminen A. Low back pain in relation to lumbar disc degeneration. Spine (Phila Pa 1976) 2000;25(4):487–492. doi: 10.1097/00007632-200002150-00016. [DOI] [PubMed] [Google Scholar]
  • 3. Patel KP, Sandy JD, Akeda K, et al. Aggrecanases and aggrecanase-generated fragments in the human intervertebral disc at early and advanced stages of disc degeneration. Spine (Phila Pa 1976) 2007;32(23):2596–2603. doi: 10.1097/BRS.0b013e318158cb85. [DOI] [PubMed] [Google Scholar]
  • 4. Beattie PF. Current understanding of lumbar intervertebral disc degeneration: a review with emphasis upon etiology, pathophysiology, and lumbar magnetic resonance imaging findings. J Orthop Sports Phys Ther. 2008;38(6):329–340. doi: 10.2519/jospt.2008.2768. [DOI] [PubMed] [Google Scholar]
  • 5. Lee JY, Lee HI, Lee SH, Kim NH. Mechanical influence of facet tropism in patients with chronic discogenic pain disorder. Bone Joint Res. 2024;13(9):452–461. doi: 10.1302/2046-3758.139.BJR-2023-0363.R1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6. Zhang GZ, Li L, Luo ZB, Zhang CY, Wang YG, Kang XW. Identification and experimental validation of key extracellular proteins as potential targets in intervertebral disc degeneration. Bone Joint Res. 2023;12(9):522–535. doi: 10.1302/2046-3758.129.BJR-2022-0369.R2. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7. Adams MA, Roughley PJ. What is intervertebral disc degeneration, and what causes it? Spine (Phila Pa 1976) 2006;31(18):2151–2161. doi: 10.1097/01.brs.0000231761.73859.2c. [DOI] [PubMed] [Google Scholar]
  • 8. Rajasekaran S, Naresh-Babu J, Murugan S. Review of postcontrast MRI studies on diffusion of human lumbar discs. J Magn Reson Imaging. 2007;25(2):410–418. doi: 10.1002/jmri.20853. [DOI] [PubMed] [Google Scholar]
  • 9. Mills ES, Talehakimi A, Urness M, et al. Anteroposterior pelvic radiograph findings correlate with sagittal spinopelvic motion. Bone Joint J. 2023;105-B(5):496–503. doi: 10.1302/0301-620X.105B5.BJJ-2022-0945.R1. [DOI] [PubMed] [Google Scholar]
  • 10. Ruiz-Fernández C, Ait Eldjoudi D, González-Rodríguez M, et al. Monomeric CRP regulates inflammatory responses in human intervertebral disc cells. Bone Joint Res. 2023;12(3):189–198. doi: 10.1302/2046-3758.123.BJR-2022-0223.R1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11. Boos N, Weissbach S, Rohrbach H, Weiler C, Spratt KF, Nerlich AG. Classification of age-related changes in lumbar intervertebral discs. Spine (Phila Pa 1976) 2002;27(23):2631–2644. doi: 10.1097/01.BRS.0000035304.27153.5B. [DOI] [PubMed] [Google Scholar]
  • 12. Rodriguez AG, Slichter CK, Acosta FL, et al. Human disc nucleus properties and vertebral endplate permeability. Spine (Phila Pa 1976) 2011;36(7):512–520. doi: 10.1097/BRS.0b013e3181f72b94. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13. Magnier C, Boiron O, Wendling-Mansuy S, Chabrand P, Deplano V. Nutrient distribution and metabolism in the intervertebral disc in the unloaded state: a parametric study. J Biomech. 2009;42(2):100–108. doi: 10.1016/j.jbiomech.2008.10.034. [DOI] [PubMed] [Google Scholar]
  • 14. Roberts S, Urban JPG, Evans H, Eisenstein SM. Transport properties of the human cartilage endplate in relation to its composition and calcification. Spine (Phila Pa 1976) 1996;21(4):415–420. doi: 10.1097/00007632-199602150-00003. [DOI] [PubMed] [Google Scholar]
  • 15. Arpinar VE, Rand SD, Klein AP, Maiman DJ, Muftuler LT. Changes in perfusion and diffusion in the endplate regions of degenerating intervertebral discs: a DCE-MRI study. Eur Spine J. 2015;24(11):2458–2467. doi: 10.1007/s00586-015-4172-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16. Cao Y, Guo Q-W, Wan Y. Significant association between the T2 values of vertebral cartilage endplates and pfirrmann grading. Orthop Surg. 2020;12(4):1164–1172. doi: 10.1111/os.12727. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17. Peters M, Willems P, Weijers R. Pseudarthrosis after lumbar spinal fusion: the role of (1)(8)F-fluoride PET/CT. Eur J Nucl Med Mol Imaging. 2015;42(12):1891–1898. doi: 10.1007/s00259-015-3154-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18. Podichetty VK. The aging spine: the role of inflammatory mediators in intervertebral disc degeneration. Cell Mol Biol (Noisy-le-grand) 2007;53(5):4–18. [PubMed] [Google Scholar]
  • 19. Watanabe A, Benneker LM, Boesch C, Watanabe T, Obata T, Anderson SE. Classification of intervertebral disk degeneration with axial T2 mapping. AJR Am J Roentgenol. 2007;189(4):936–942. doi: 10.2214/AJR.07.2142. [DOI] [PubMed] [Google Scholar]
  • 20. Czernin J, Satyamurthy N, Schiepers C. Molecular mechanisms of bone 18F-NaF deposition. J Nucl Med. 2010;51(12):1826–1829. doi: 10.2967/jnumed.110.077933. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21. Piert M, Zittel TT, Becker GA, et al. Assessment of porcine bone metabolism by dynamic. J Nucl Med. 2001;42(7):1091–1100. [PubMed] [Google Scholar]
  • 22. Peters MJM, Wierts R, Jutten EMC, Halders S, Willems P, Brans B. Evaluation of a short dynamic 18F-fluoride PET/CT scanning method to assess bone metabolic activity in spinal orthopedics. Ann Nucl Med. 2015;29(9):799–809. doi: 10.1007/s12149-015-1008-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23. Kogan F, Fan AP, Monu U, Iagaru A, Hargreaves BA, Gold GE. Quantitative imaging of bone-cartilage interactions in ACL-injured patients with PET-MRI. Osteoarthritis Cartilage. 2018;26(6):790–796. doi: 10.1016/j.joca.2018.04.001. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24. Kirchner F, Pinar A, Milani I, Prado R, Padilla S, Anitua E. Vertebral intraosseous plasma rich in growth factor (PRGF-Endoret) infiltrations as a novel strategy for the treatment of degenerative lesions of endplate in lumbar pathology: description of technique and case presentation. J Orthop Surg Res. 2020;15(1):72. doi: 10.1186/s13018-020-01605-w. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25. Buchbender C, Ostendorf B, Ruhlmann V, et al. Hybrid 18F-labeled fluoride positron emission tomography/magnetic resonance (MR) imaging of the sacroiliac joints and the spine in patients with axial spondyloarthritis: a pilot study exploring the link of MR bone pathologies and increased osteoblastic activity. J Rheumatol. 2015;42(9):1631–1637. doi: 10.3899/jrheum.150250. [DOI] [PubMed] [Google Scholar]
  • 26. Pfirrmann CWA, Metzdorf A, Zanetti M, Hodler J, Boos N. Magnetic resonance classification of lumbar intervertebral disc degeneration. Spine (Phila Pa 1976) 2001;26(17):1873–1878. doi: 10.1097/00007632-200109010-00011. [DOI] [PubMed] [Google Scholar]
  • 27. Weng Y-S, Tang C-T, Chang W-C, et al. Added value of 18Fluorine-fluorodeoxyglucose (18F-FDG) PET/MRI for evaluation of failed back surgery syndrome: comparison with non-contrast MRI. Jpn J Radiol. 2025;43(3):509–519. doi: 10.1007/s11604-024-01679-0. [DOI] [PubMed] [Google Scholar]
  • 28. Raj PP. Intervertebral disc: anatomy-physiology-pathophysiology-treatment. Pain Pract. 2008;8(1):18–44. doi: 10.1111/j.1533-2500.2007.00171.x. [DOI] [PubMed] [Google Scholar]
  • 29. Mamisch TC, Hughes T, Mosher TJ, et al. T2 star relaxation times for assessment of articular cartilage at 3 T: a feasibility study. Skeletal Radiol. 2012;41(3):287–292. doi: 10.1007/s00256-011-1171-x. [DOI] [PubMed] [Google Scholar]
  • 30. Dunn TC, Lu Y, Jin H, Ries MD, Majumdar S. T2 relaxation time of cartilage at MR imaging: comparison with severity of knee osteoarthritis. Radiology. 2004;232(2):592–598. doi: 10.1148/radiol.2322030976. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31. Tanaka M. Biomechanical investigation on the influence of the regional material degeneration of an intervertebral disc in a lower lumbar spinal unit: a finite element study. Comput Biol Med. 2018;98:26–38. doi: 10.1016/j.compbiomed.2018.05.010. [DOI] [PubMed] [Google Scholar]
  • 32. Shahraki NM, Fatemi A, Agarwal A, Goel VK. Prediction of clinically relevant initiation and progression of tears within annulus fibrosus. J Orthop Res. 2017;35(1):113–122. doi: 10.1002/jor.23346. [DOI] [PubMed] [Google Scholar]
  • 33. Ma M, Zhang C, Zhong Z, et al. siRNA incorporated in slow-release injectable hydrogel continuously silences DDIT4 and regulates nucleus pulposus cell pyroptosis through the ROS/TXNIP/NLRP3 axis to alleviate intervertebral disc degeneration. Bone Joint Res. 2024;13(5):247–260. doi: 10.1302/2046-3758.135.BJR-2023-0320.R1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34. Lyons G, Eisenstein SM, Sweet MB. Biochemical changes in intervertebral disc degeneration. Biochim Biophys Acta. 1981;673(4):443–453. doi: 10.1016/0304-4165(81)90476-1. [DOI] [PubMed] [Google Scholar]
  • 35. Canbay S, Turhan N, Bozkurt M, Arda K, Caglar S. Correlation of matrix metalloproteinase-3 expression with patient age, magnetic resonance imaging and histopathological grade in lumbar disc degeneration. Turk Neurosurg. 2013;23(4):427–433. doi: 10.5137/1019-5149.JTN.7459-12.0. [DOI] [PubMed] [Google Scholar]
  • 36. Erasmus JJ, Gladish GW, Broemeling L, et al. Interobserver and intraobserver variability in measurement of non-small-cell carcinoma lung lesions: implications for assessment of tumor response. J Clin Oncol. 2003;21(13):2574–2582. doi: 10.1200/JCO.2003.01.144. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Data Availability Statement

The datasets generated and analyzed in the current study are not publicly available due to data protection regulations. Access to data is limited to the researchers who have obtained permission for data processing. Further inquiries can be made to the corresponding author.

Articles from Bone & Joint Research are provided here courtesy of British Editorial Society of Bone and Joint Surgery

RESOURCES