Skip to main content
NCBI home page
Acta Radiologica Open logoLink to Acta Radiologica Open
. 2025 Sep 25;14(9):20584601251379482. doi: 10.1177/20584601251379482

Progression of lumbar disc degeneration: A 14-year follow-up study examining Pfirrmann grading and its individual disc components

Niko Murto 1,✉, Teija Lund 2, Hannu Kautiainen 3,4, Katariina Luoma 1, Liisa Kerttula 1
PMCID: PMC12475308  PMID: 41018258

Abstract

Background

Lumbar disc degeneration (LDD) is frequently evaluated using the Pfirrmann classification. While this composite grading system provides an overview of degeneration severity, it may oversimplify LDD by overlooking variability in individual disc components, reducing its effectiveness in longitudinal studies and constraining its applicability in artificial intelligence-based image analysis.

Purpose

To examine the 14-year progression of LDD using the Pfirrmann classification and its individual components, and to evaluate the potential of component-based analysis.

Material and Methods

LDD was assessed using MRI in 19 males (95 discs) at ages 37 and 51 by two radiologists. Evaluations included Pfirrmann grading, quantitative nucleus pulposus (NP) signal intensity, and visual grading of NP inhomogeneity, annulus fibrosus (AF) border distinction, and disc height (DH). Analyses included longitudinal changes in LDD variables and correlations between Pfirrmann grading and disc components. To assess overall LDD, a summary score was calculated by summing individual disc grades.

Results

Pfirrmann grading correlated strongly with AF border distinction, moderately with NP signal intensity, and weakly with NP inhomogeneity and DH. Pfirrmann summary score (range 5–25) increased by 3 points over time. Variability was observed in the progression of individual disc component degeneration. While mean NP signal intensity significantly decreased, some discs exhibited increase.

Conclusion

This longitudinal study highlights complexity of LDD and variability in disc component changes. While Pfirrmann classification captures overall degeneration, its limitations in detecting subtle variations in disc components suggest a need for more detailed assessments to enhance diagnostic precision and support the development of automated analysis tools.

Keywords: lumbar disc degeneration, Pfirrmann, longitudinal study, nucleus pulposus, annulus fibrosus, disc height

Introduction

The clinical relevance of lumbar disc degeneration (LDD) observed on magnetic resonance imaging (MRI) remains a topic of ongoing debate. Degenerative changes are frequently detected in intervertebral discs of asymptomatic individuals, 1 making it challenging to differentiate between normal age-related degeneration and pathological LDD that may contribute to low back pain (LBP). 2 The severity of LDD is often assessed using the Pfirrmann classification. 3 While advanced degeneration, as classified by Pfirrmann grading, has been associated with LBP in some studies,4,5 findings remain inconsistent. Other studies have found no significant correlation between Pfirrmann grades and pain6,7 suggesting limitations in the classification’s ability to distinguish clinically relevant changes.

The Pfirrmann classification provides a comprehensive summary of LDD by evaluating four key characteristics: signal intensity and homogeneity of nucleus pulposus (NP), distinction between NP and annulus fibrosus (AF), and disc height (DH), categorizing LDD into five grades. 3 While widely recognized for its fair to excellent inter- and intra-observer reliability3,8,9 and its simplicity and ease of application, variations in visual perception—particularly when assessing subtle changes—can lead to discrepancies between observers, resulting in inconsistencies at grade boundaries.10,11

Furthermore, the Pfirrmann classification has been criticized for its limited discriminatory power in elderly populations, potentially reducing its accuracy in tracking LDD progression over time. 12 Additionally, it fails to detect early biochemical alterations, as it relies solely on morphological features observable on conventional MRI.13–15 As a qualitative classification, it lacks quantitative data, limiting its sensitivity to subtle degenerative changes over time. Additionally, its lack of specificity for disc components reduces its ability to assess localized degeneration. Without evaluating specific structural components, longitudinal studies may overlook regional variations, leading to an incomplete understanding of LDD progression.

To address these limitations, various modifications to the Pfirrmann classification11,12,16 and quantitative methods17–21 have been proposed to improve the accuracy and objectivity of LDD assessment.

To our knowledge, no previous longitudinal study has independently analyzed each component of the Pfirrmann classification. This study aims to 1 : examine the 14-year progression of degenerative changes in specific intervertebral disc components included in the Pfirrmann grading system, and 2 assess longitudinal changes in quantitative nucleus pulposus (NP) signal intensity. By focusing on individual structural features rather than the composite score alone, this study seeks to generate insights that support the development of more precise, artificial intelligence (AI)-driven tools for structural disc assessment.

Materials and methods

This study examines a subset of 75 individuals originally recruited at age 20 for a longitudinal investigation into lumbar disc degeneration (LDD). Participants were selected during their national military service due to a history of low back pain (LBP) severe enough to interfere with their ability to serve. These individuals were not professional soldiers, but rather conscripts fulfilling their mandatory military service. Approval was obtained from the ethics committee of Helsinki University Hospital (reference number HUS/1493/2017).

Lumbar spine magnetic resonance imaging (MRI) examinations were conducted at two time points, ages 37 and 51. All 75 original participants were invited by mail to attend the MRI scans; however, six individuals could not be contacted. Of the remaining 69, 32 completed MRI scans at age 37, and 19 participated in both the 37- and 51-year-old MRI assessments. This study focuses on these 19 individuals, forming the cohort for this 14-year follow-up.

Routine T2-weighted and T1-weighted sagittal sequences, along with T2-weighted axial sequences, were acquired. For assessing LDD, all T2-weighted sagittal images from both time points were analyzed. Imaging at age 37 was performed using a 1.0 T GE Signa scanner with the following parameters for T2-weighted sagittal sequences: TR/TE (repetition time/echo time) 4000/97 ms, field of view 260 × 260 mm, and matrix size 256 × 256. Imaging at age 51 was conducted using a 1.5 T Siemens Healthineers Avanto Fit scanner, with parameters for T2-weighted sagittal sequences as follows: TR/TE 3520/117 ms, field of view 300 × 300 mm and matrix size 326 × 384.

Patient-reported outcomes, including low back pain (measured using the Visual Analogue Scale [VAS]) and functional status (assessed with the Low Back Outcome Score [LBOS]), were collected at both imaging time points.

Two radiologists—one a musculoskeletal radiology fellow (first author) and the other a specialist in musculoskeletal radiology (fifth author)—independently evaluated 95 discs. Both radiologists were blinded to each other’s assessments. The evaluation included Pfirrmann grading and visual grading of nucleus pulposus (NP) inhomogeneity, annulus fibrosus (AF) border distinction and disc height (DH) 3 (Table 1). Additionally, a previously published targeted quantitative method was used to measure nucleus pulposus (NP) signal intensity, focusing on the brightest and most homogeneous area of the NP. 21 The NP signal intensity values were individually adjusted for cerebrospinal fluid signal intensity. The MRI scans at ages 37 and 51 were assessed independently and separately.

Table 1.

Visual grading criteria for nucleus pulposus (NP) inhomogeneity, annulus fibrosus (AF) border distinction and disc height. 3

graphic file with name 10.1177_20584601251379482-img1.jpg
Open in a new tab

Inter-rater agreement between the two radiologists was assessed. Summary scores were calculated by summing the individual grades for each of the five lumbar intervertebral discs (L1/2 to L5/S1) across the evaluated variables (Pfirrmann grading, NP signal intensity, NP inhomogeneity, AF border distinction and DH). This approach has been used in previous studies22–26 to provide a comprehensive representation of degenerative changes in the lumbar spine for statistical analysis.

NP inhomogeneity, AF border distinction, and DH are subcomponents of the Pfirrmann classification, providing a more detailed assessment of LDD within disc components. Quantitative NP signal intensity was included as a control variable to objectively assess changes in NP degeneration, complementing the qualitative evaluations. The progression of these variables over 14 years was analyzed and their correlations with changes in Pfirrmann grade were assessed across individual discs and summary scores.

Statistics

Data are presented as means with standard deviation (SD) or as counts (n) with percentages (%). Mean changes (within subjects) were made by using bootstrap type paired t test. Correlations are expressed with Spearman’s correlation coefficients with bootstrapped (10,000 replications) 95% confidence intervals. Spearman’s correlations values above 0.20, 0.40 and 0.60 represent weak, moderate and strong relationships, respectively. 27 Inter-rater agreement was estimated by using type of accuracy (overall correctly classified, percentage) and the kappa statistic (κ) with ordinal weights. 95% CIs for estimates were obtained using bias-corrected bootstrap with 10,000 iterations. The normality of variables was evaluated graphically and by using the Shapiro-Wilk W-test. Stata 18.0 (StataCorp LP; College Station, Texas, USA) statistical package was used for the analysis.

Results

Inter-rater agreement between the radiologists was very good for nucleus pulposus (NP) inhomogeneity and annulus fibrosus (AF) border distinction. Agreement for disc height (DH) was moderate (Table 2). For Pfirrmann grading, agreement was almost perfect (κ = 0.89, 95% CI: 0.84–0.93). The repeatability of NP signal intensity measurements was previously reported as excellent, with an intraclass correlation coefficient (ICC) of 0.94–0.99. 21 Given the strong inter-rater agreement for all variables except DH, the values from the more experienced rater (fifth author) were used in the analysis, and no consensus ratings were performed.

Table 2.

Inter-rater agreement (Cohen’s κ and percentage agreement) for nucleus pulposus (NP) inhomogeneity, annulus fibrosus (AF) border distinction, and disc height.

Kappa a (95% CI) Agreement %, (95% CI)
NP inhomogeneity 0.87 (0.77 to 0.96) 98 (96 to 100)
AF border distinction 0.90 (0.84 to 0.96) 96 (94 to 98)
Disc height 0.42 (0.32 to 0.51) 87 (85 to 89)
Open in a new tab

aKappa statistic (κ) with ordinal weights.

At age 37, most discs were graded Pfirrmann grade 2 (n = 59, 62%), with fewer in grade 3 (n = 16, 17%), grade 4 (n = 18, 19%), and grade 5 (n = 2, 2%). By age 51, the proportion of grade 2 discs had decreased (n = 25, 26%), while the majority had progressed to grade 3 (n = 32, 34%) or grade 4 (n = 31, 33%), with a few reaching grade 5 (n = 7, 7%). No grade 1 discs were observed at either age 37 or 51.

The AF border distinction showed the highest proportion of normal findings (grade 0) at age 37 and exhibited the most frequent progression over 14 years, followed by a reduction in DH. NP inhomogeneity had the fewest normal discs at age 37, and progression in NP inhomogeneity was less common (Table 3).

Table 3.

Changes in nucleus pulposus (NP) inhomogeneity, annulus fibrosus (AF) border distinction, and disc height during follow-up (−1 = improvement by one grade, 0 = no change, 1 = progression by one grade, 2 = progression by two grades etc.) across all 95 discs.

NP inhomogeneity change NP inhomogeneity at age 37 Total
Homogeneous, uniformly bright or gray (0) Inhomogeneous, speckled or patchy structure (1) Markedly inhomogeneous, uniformly dark or black (2)
−1 0 0 2 2
0 0 57 23 80
1 1 12 0 13
Total 1 69 28 95
Open in a new tab

Changes in the Pfirrmann grade were strongly correlated with changes in AF border distinction (r = 0.68, 95% CI: 0.55–0.77) and moderately correlated with measured NP signal intensity (r = −0.40, 95% CI: −0.55 to −0.21). Weaker correlations were observed between changes in the Pfirrmann grade and changes in NP inhomogeneity (r = 0.38, 95% CI: 0.19–0.54) as well as DH (r = 0.22, 95% CI: 0.01–0.40) (Table 4).

Table 4.

The Spearman correlation coefficients between changes in the Pfirrmann grade and nucleus pulposus (NP) signal intensity, NP inhomogeneity, annulus fibrosus (AF) border distinction and disc height during follow-up (A) across all 95 discs (representing all lumbar discs from the subjects) and (B) the summary scores (calculated as the total scores of the five lumbar discs for each subject).

Correlation with Pfirrmann change (n = 95) (r) 95% confidence interval (CI)
A
NP signal intensity change (n = 95) −0.40 −0.55 to −0.21
NP inhomogeneity change (n = 95) 0.38 0.19 to 0.54
AF border distinction change (n = 95) 0.68 0.55 to 0.77
Disc height change (n = 95) 0.22 0.01 to 0.40
Open in a new tab

Over the 14-year period, the Pfirrmann summary score increased significantly by 3 points (95% CI: 2.1 to 3.9), rising from a mean score of 13.1 (SD 1.5) at age 37 to 16.1 (SD 2.1) at age 51 (p < .001). A strong correlation was observed between changes in the Pfirrmann summary score and the AF border distinction summary score (r = 0.67, 95% CI: 0.31 to 0.86), while the correlation with the NP signal intensity summary score was moderate (r = −0.57, 95% CI: −0.81 to −0.15). In contrast, weaker correlations were found between the Pfirrmann summary score and the NP inhomogeneity summary score (r = 0.21, 95% CI: −0.27 to 0.60) and a very weak correlation with the DH summary score (r = 0.04, 95% CI: −0.42 to 0.48).

The NP signal intensity summary score across the lumbar spine significantly decreased by 0.43 points (95% CI: −0.73 to −0.03), from 2.08 (SD 0.24) at age 37 to 1.50 (SD 0.24) at age 51 (p < .001). Despite the overall decrease in mean NP signal intensity, some individual discs showed an increase in NP signal intensity, a trend observed only in the lower lumbar levels and in discs with initially lower NP signal intensity at age 37 (Figure 1). These signal variations contributed to inconsistencies in NP inhomogeneity and AF border distinction assessments between ages 37 and 51 (Figure 2). In contrast, the discs with the brightest NP signal intensity at age 37 exhibited the greatest decrease in NP signal intensity (Figure 1).

Figure 1.

Figure 1.

Open in a new tab

Scatterplot of nucleus pulposus signal intensity change over the 14-year follow-up by nucleus pulposus signal intensity at age 37.

Figure 2.

Figure 2.

Open in a new tab

Illustration of discrepancies in degenerative changes during follow-up in two subjects (a) and (b). The L4/5 disc in both subjects shows increased nucleus pulposus (NP) signal intensity at age 51 compared to 37, seen as a bright focus in the NP—referred to as an intranuclear high-intensity zone. This increased signal is accompanied by a more distinct separation between the NP and the annulus fibrosus and reduced disc height over time.

Statistical analysis revealed no significant relationships between patient-reported outcomes and either the Pfirrmann grade or changes in disc components. At age 37, 8 of the 19 participants (42%) reported moderate to severe low back pain (VAS ≥4). The same number of participants reported moderate to severe pain at age 51. Of these, 6 individuals (32%) experienced this level of pain at both time points.

Discussion

Our findings confirm the expected progression of lumbar disc degeneration (LDD), as evidenced by a significant increase in the Pfirrmann summary score and a decrease in the nucleus pulposus (NP) signal intensity summary score. However, the weak correlations of NP inhomogeneity and disc height (DH) with Pfirrmann grading, along with variability in changes of NP inhomogeneity, annulus fibrosus (AF) border distinction, and DH during follow-up, as well as the increase in NP signal intensity in specific discs, highlight the challenges in applying the Pfirrmann classification in a straightforward manner in the longitudinal assessment of LDD.

Despite variability in LDD findings across individuals of the same age, a general linear progression of LDD with aging has been observed.28,29 Previous studies have demonstrated a strong correlation between Pfirrmann grades and aging, with discs typically progressing from lower to higher grades over decades. For instance, Pfirrmann grade 3 discs are most prevalent in the 2nd to 5th decades, while grade 4 discs become more common after the 6th decade. 30

LDD is characterized by a reduction in proteoglycan content in the NP, which leads to diminished water retention capacity, reflected as decreased T2 signal intensity on MRI. Additionally, changes in collagen composition and increased fibrosis in the NP further impair water retention and signal characteristics, resulting in the observed inhomogeneity on T2-weighted images.14,29 Moreover, the conversion of collagen type I to type II in the inner AF, along with inward folding of the inner AF, obscures the distinction between the NP and AF. 28

The strong correlation observed between changes in the Pfirrmann grade and changes in AF border distinction (r = 0.68) suggests that AF border could be a valuable indicator of LDD in longitudinal studies within the age group examined in this study. In contrast, the weak correlations between changes in the Pfirrmann grade and changes in NP inhomogeneity (r = 0.38) and DH (r = 0.22) highlight the variability in the degeneration of disc components. These weaker associations indicate that changes in these disc components may progress at a different rate than overall changes in Pfirrmann grade, emphasizing the limitations of the Pfirrmann classification in capturing all aspects of disc pathology.

A gradual reduction in DH is considered a normal part of the aging process. Both genetic and environmental factors significantly influence the individual progression of DH reduction 31 ; however, excessive or repetitive mechanical overload, such as trauma, can accelerate DH reduction by causing disc damage. 32 While a collapsed disc is often viewed as the final stage of LDD, DH is considered an unreliable indicator of early-stage LDD. 33 The relationship between Pfirrmann grade and DH is nonlinear, with significant reduction in DH occurring only at the most advanced stage of degeneration (Pfirrmann grade 5). In contrast, DH remains relatively stable across grades 1 to 4,34,35 highlighting the limitations of using DH as the sole indicator of LDD progression, especially in its early and intermediate stages.

The visual grading of DH presents challenges, as determining what constitutes a “normal” DH is subjective and often ambiguous. Our inter-rater agreement between radiologists for DH was weaker compared to other variables (κ = 0.42). This underscores the importance of using quantitative DH measurements, 34 particularly in longitudinal studies, to ensure consistency and objectivity. In a previous study, we reported excellent repeatability of NP signal intensity measurements, 21 further supporting the reliability of quantitative assessments in LDD research.

The inverse correlation between changes in measured NP signal intensity and Pfirrmann grade (r = −0.40) reflects the expected decrease in NP proteoglycan and water content during LDD progression. Notably, the observed increase in NP signal intensity during follow-up—seen in only a few specific discs as a focal bright area previously referred to as an intranuclear high-intensity zone 21 —likely represents intradiscal fluid accumulation. This finding may reflect fluid replacing gas within intradiscal vacuum phenomena, a process associated with advanced disc degeneration, supine imaging position, and degenerative endplate changes such as Modic type 1 lesions.36–39 In contrast, spinal infection may also present with increased NP signal intensity; however, spondylodiscitis typically manifests with more diffuse T2 hyperintensity extending throughout the disc, accompanied by adjacent endplate marrow and paraspinal soft tissue oedema, and clinical signs of infection such as pain, fever, and elevated inflammatory markers. 40

This increase in NP signal intensity poses challenges for LDD grading, especially in longitudinal studies, as it complicates the assessment of Pfirrmann grades and the evaluation of both NP and AF. While NP signal intensity typically decreases with advancing DD, occasional increases due to fluid accumulation may enhance AF border distinction and complicate the grading of NP inhomogeneity. Since Pfirrmann grading primarily relies on NP signal intensity and structural features of NP and AF, a temporary increase in signal intensity may lead to misinterpretations, such as underestimating degeneration severity or incorrectly assuming reversal of LDD. Consequently, these variations may result in inconsistent conclusions regarding degenerative changes in longitudinal studies.

Certain limitations of this study should be acknowledged. The small sample size, resulting from participant dropout over the 14-year follow-up period, limit the generalisability of the findings. Despite multiple invitations, some individuals could not be reached or declined participation, reducing the number of available follow-up magnetic resonance images. This may have affected the study’s ability to fully capture the broader variability in LDD progression seen in larger populations. The absence of statistically significant relationships between patient-reported outcomes and imaging findings is likely attributable to the limited sample size. Furthermore, all participants had a history of low back pain (LBP) beginning in early adulthood, often with a chronic or fluctuating course. As a result, establishing associations between structural changes and symptoms is inherently challenging in such a pre-selected symptomatic population. Additionally, the homogeneity of our study cohort in terms of age and sex limits the applicability of our results to other populations, as the observed LDD progression may be specific to men aged 37–51.

Furthermore, while our study focuses on the Pfirrmann classification and its components, we did not assess other markers of degeneration, such as endplate changes and Modic changes, which have been associated with LBP. 41 Including these degenerative markers, alongside the evaluation of disc changes, could provide additional insights and help differentiate age-related LDD from pathological LDD.

We also did not incorporate more advanced imaging techniques in the assessment of LDD. Quantitative MRI methods, such as T2 mapping, T1 relaxation time and T1rho imaging,19,42 could provide deeper insights into the biochemical changes occurring within disc tissues and serve as earlier indicators of degeneration compared to conventional magnetic resonance imaging (MRI). These approaches may better capture the continuum of LDD, addressing the limitations of categorical classifications. However, these techniques require specialized MRI sequences that are not part of routine clinical imaging, resulting in reduced accessibility, longer scan times, and increased susceptibility to artefacts.43,44

Despite the limitations outlined, the longitudinal design of this study and its comprehensive assessment of structural components contribute to understanding the complexity of LDD.

In summary, our study demonstrated significant progression of LDD over 14 years, reflected by an increase in the Pfirrmann summary score and a decrease in the NP signal intensity summary score. However, the progression of individual disc components varied, with AF border showing the strongest correlation with Pfirrmann grading, while NP inhomogeneity and disc height displayed weaker associations. Notably, some discs exhibited an increase in NP signal intensity over time, resulting in discrepancies in NP inhomogeneity and AF border distinction assessments.

These findings underscore the limitations of the Pfirrmann classification in capturing the heterogeneous and multifaceted nature of LDD progression, particularly in longitudinal studies. While the Pfirrmann classification remains a useful framework for evaluating overall LDD severity, its broad categorization may obscure subtle structural changes. This is especially relevant in the context of AI-based classification tools, where Pfirrmann grading has become one of the most commonly used methods for LDD classification. 45

In conclusion, our results highlight the value of analyzing individual disc components separately, as they progress at different rates and reflect distinct structural changes. Relying solely on composite Pfirrmann grades may limit the accuracy of AI models. Future research should focus on refining grading criteria and incorporating quantitative imaging techniques to enhance the precision of LDD assessment and better characterize the progression of degeneration over time.

Footnotes

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding: The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: First author received grants from the Helsinki University Hospital research fund (Y780021010, Y780020126, Y7800MILJO and Y780023030) and from Musculoskeletalradiologists of Finland association, The Radiological Society of Finland and Pehr Oscar Klingendahl Fund.

Ethics approval

This study was performed in line with the principles of the Declaration of Helsinki. Approval was obtained from the ethics committee of Helsinki University Hospital (reference number HUS/1493/2017).

Consent to participate

Informed consent was obtained from all individual participants included in the study.

Consent for publication

Patients signed informed consent regarding the publishing of their data.

ORCID iDs

Niko Murto https://orcid.org/0000-0002-9820-1390

Liisa Kerttula https://orcid.org/0009-0006-6907-8094

Data Availability Statement

The data that support the findings of this study are available from Helsinki University Hospital, but restrictions apply to the availability of these data, which were used under license for the current study, and so are not publicly available. Data are, however, available from the authors upon reasonable request and with permission of Helsinki University Hospital.*

References

  • 1.Brinjikji W, Luetmer PH, Comstock B, et al. Systematic literature review of imaging features of spinal degeneration in asymptomatic populations. AJNR Am J Neuroradiol 2015; 36: 811–816. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Oei MW, Evens AL, Bhatt AA, et al. Imaging of the aging spine. Radiol Clin 2022; 60: 629–640. [DOI] [PubMed] [Google Scholar]
  • 3.Pfirrmann CW, Metzdorf A, Zanetti M, et al. Magnetic resonance classification of lumbar intervertebral disc degeneration. Spine 2001; 26: 1873–1878. [DOI] [PubMed] [Google Scholar]
  • 4.Brinjikji W, Diehn FE, Jarvik JG, et al. MRI findings of disc degeneration are more prevalent in adults with low back pain than in asymptomatic controls: a systematic review and meta-analysis. AJNR Am J Neuroradiol 2015; 36: 2394–2399. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Smith A, Hancock M, O’Hanlon S, et al. The association between different trajectories of low back pain and degenerative imaging findings in young adult participants within the raine study. Spine 2022; 47: 269–276. [DOI] [PubMed] [Google Scholar]
  • 6.Kasch R, Truthmann J, Hancock MJ, et al. Association of lumbar MRI findings with current and future back pain in a population-based cohort study. Spine 2022; 47: 201–211. [DOI] [PubMed] [Google Scholar]
  • 7.Steffens D, Hancock MJ, Maher CG, et al. Does magnetic resonance imaging predict future low back pain? A systematic review. Eur J Pain 2014; 18: 755–765. [DOI] [PubMed] [Google Scholar]
  • 8.Urrutia J, Besa P, Campos M, et al. The Pfirrmann classification of lumbar intervertebral disc degeneration: an independent inter- and intra-observer agreement assessment. Eur Spine J; 25: 2728. [DOI] [PubMed] [Google Scholar]
  • 9.Carrino JA, Lurie JD, Tosteson ANA, et al. Lumbar spine: reliability of MR imaging findings. Radiology 2009; 250: 161–170. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Nagy SA, Juhasz I, Komaromy H, et al. A statistical model for intervertebral disc degeneration: determination of the optimal T2 cut-off values. Clin Neuroradiol 2014; 24: 355–363. [DOI] [PubMed] [Google Scholar]
  • 11.Soydan Z, Bayramoglu E, Urut DU, et al. Tracing the disc: the novel qualitative morphometric MRI based disc degeneration classification system. JOR Spine 2024; 7: e1321. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Griffith JF, Wang Y-XJ, Antonio GE, et al. Modified Pfirrmann grading system for lumbar intervertebral disc degeneration. Spine 2007; 32: E708–E712. [DOI] [PubMed] [Google Scholar]
  • 13.Samartzis D, Borthakur A, Belfer I, et al. Novel diagnostic and prognostic methods for disc degeneration and low back pain. Spine J 2015; 15: 1919–1932. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Benneker LM, Heini PF, Anderson SE, et al. Correlation of radiographic and MRI parameters to morphological and biochemical assessment of intervertebral disc degeneration. Eur Spine J 2005; 14: 27–35. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Auerbach JD, Johannessen W, Borthakur A, et al. In vivo quantification of human lumbar disc degeneration using T(1rho)-weighted magnetic resonance imaging. Eur Spine J 2006; 15 Suppl 3(Suppl 3): S338–S344. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Noebauer-Huhmann I-M, Juras V, Pfirrmann CWA, et al. Sodium MR imaging of the lumbar intervertebral disk at 7 T: correlation with T2 mapping and modified Pfirrmann score at 3 T--preliminary results. Radiology 2012; 265: 555–564. [DOI] [PubMed] [Google Scholar]
  • 17.Rim DC. Quantitative Pfirrmann disc degeneration grading system to overcome the limitation of Pfirrmann disc degeneration grade. Korean J Spine 2016; 13: 1–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Li Z, Wu Y, Liu M, et al. A novel quantitative method to evaluate lumbar disc degeneration: MRI histogram analysis. Eur Spine J 2024; 33: 2420–2429. [DOI] [PubMed] [Google Scholar]
  • 19.Yoon MA, Hong S-J, Kang CH, et al. T1rho and T2 mapping of lumbar intervertebral disc: correlation with degeneration and morphologic changes in different disc regions. Magn Reson Imaging 2016; 34: 932–939. [DOI] [PubMed] [Google Scholar]
  • 20.Stelzeneder D, Welsch GH, Kovács BK, et al. Quantitative T2 evaluation at 3.0 T compared to morphological grading of the lumbar intervertebral disc: a standardized evaluation approach in patients with low back pain. Eur J Radiol 2012; 81: 324–330. [DOI] [PubMed] [Google Scholar]
  • 21.Murto N, Luoma K, Lund T, et al. Reliability of T2-weighted signal intensity-based quantitative measurements and visual grading of lumbar disc degeneration on MRI. Acta Radiol. 2023; 64: 2145–2151. [DOI] [PubMed] [Google Scholar]
  • 22.Mok FPS, Samartzis D, Karppinen J, et al. ISSLS prize winner: prevalence, determinants, and association of Schmorl nodes of the lumbar spine with disc degeneration: a population-based study of 2449 individuals. Spine 2010; 35: 1944–1952. [DOI] [PubMed] [Google Scholar]
  • 23.Määttä JH, Karppinen JI, Luk KDK, et al. Phenotype profiling of Modic changes of the lumbar spine and its association with other MRI phenotypes: a large-scale population-based study. Spine J 2015; 15: 1933–1942. [DOI] [PubMed] [Google Scholar]
  • 24.Aavikko A, Lohman M, Ristolainen L, et al. ISSLS prize in clinical science 2022: accelerated disc degeneration after pubertal growth spurt differentiates adults with low back pain from their asymptomatic peers. Eur Spine J 2022; 31: 1080–1087. [DOI] [PubMed] [Google Scholar]
  • 25.Cheung KMC, Karppinen J, Chan D, et al. Prevalence and pattern of lumbar magnetic resonance imaging changes in a population study of one thousand forty-three individuals. Spine 2009; 34: 934–940. [DOI] [PubMed] [Google Scholar]
  • 26.Samartzis D, Karppinen J, Mok F, et al. A population-based study of juvenile disc degeneration and its association with overweight and obesity, low back pain, and diminished functional status. J Bone Joint Surg Am 2011; 93: 662–670. [DOI] [PubMed] [Google Scholar]
  • 27.Evans JD. Straightforward statistics for the behavioral sciences. Belmont, CA: Thomson Brooks/Cole Publishing Co, 1996. [Google Scholar]
  • 28.Adams MA, Roughley PJ. What is intervertebral disc degeneration, and what causes it? Spine 2006; 31: 2151–2161. [DOI] [PubMed] [Google Scholar]
  • 29.Boos N, Weissbach S, Rohrbach H, et al. Classification of age-related changes in lumbar intervertebral discs: 2002 Volvo Award in basic science. Spine (Phila Pa 1976) 2002; 27: 2631–2644. [DOI] [PubMed] [Google Scholar]
  • 30.Oh CH, Yoon SH. Whole spine disc degeneration survey according to the ages and sex using Pfirrmann disc degeneration grades. Korean J Spine 2017; 14: 148–154. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Karchevskaya AE, Poluektov YM, Korolishin VA. Understanding intervertebral disc degeneration: background factors and the role of initial injury. Biomedicines 2023; 11: 2714. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Stokes IAF, Iatridis JC. Mechanical conditions that accelerate intervertebral disc degeneration: overload versus immobilization. Spine 2004; 29: 2724–2732. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Frobin W, Brinckmann P, Kramer M, et al. Height of lumbar discs measured from radiographs compared with degeneration and height classified from MR images. Eur Radiol 2001; 11: 263–269. [DOI] [PubMed] [Google Scholar]
  • 34.Salamat S, Hutchings J, Kwong C, et al. The relationship between quantitative measures of disc height and disc signal intensity with Pfirrmann score of disc degeneration. SpringerPlus 2016; 5: 829. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Teichtahl AJ, Urquhart DM, Wang Y, et al. A Dose-response relationship between severity of disc degeneration and intervertebral disc height in the lumbosacral spine. Arthritis Res Ther 2015; 17: 297. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.D’Anastasi M, Birkenmaier C, Schmidt GP, et al. Correlation between vacuum phenomenon on CT and fluid on MRI in degenerative disks. AJR Am J Roentgenol 2011; 197: 1182–1189. [DOI] [PubMed] [Google Scholar]
  • 37.Cawley DT, McDonnell A, Simpkin A, et al. Intra-discal vacuum phenomenon with advanced lumbar spine disc degeneration: complementary findings from both MRI and CT. BMC Med Imag 2025; 25: 94. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Wang H-J, Chen B-B, Yu C-W, et al. Alteration of disc vacuum contents during prolonged supine positioning: evaluation with MR image. Spine 2007; 32: 2610–2615. [DOI] [PubMed] [Google Scholar]
  • 39.Schweitzer ME, El-Noueam KI. Vacuum disc: frequency of high signal intensity on T2-weighted MR images. Skelet Radiol 1998; 27: 83–86. [DOI] [PubMed] [Google Scholar]
  • 40.Shah LM, Ross JS. Imaging of degenerative and infectious conditions of the spine. Neurosurgery 2016; 79: 315–335. [DOI] [PubMed] [Google Scholar]
  • 41.van der Graaf JW, Kroeze RJ, Buckens CFM, et al. MRI image features with an evident relation to low back pain: a narrative review. Eur Spine J 2023; 32: 1830–1841. [DOI] [PubMed] [Google Scholar]
  • 42.Hwang D, Kim S, Abeydeera NA, et al. Quantitative magnetic resonance imaging of the lumbar intervertebral discs. Quant Imag Med Surg 2016; 6: 744–755. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43.Yang L, Sun C, Gong T, et al. T1ρ, T2 and T2* mapping of lumbar intervertebral disc degeneration: a comparison study. BMC Muscoskelet Disord 2022; 23: 1135. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44.Tamagawa S, Sakai D, Nojiri H, et al. Imaging evaluation of intervertebral disc degeneration and painful discs-advances and challenges in quantitative MRI. Diagnostics; 12: 707. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45.Compte R, Granville Smith I, Isaac A, et al. Are current machine learning applications comparable to radiologist classification of degenerate and herniated discs and Modic change? A systematic review and meta-analysis. Eur Spine J 2023; 32: 3764–3787. [DOI] [PMC free article] [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 data that support the findings of this study are available from Helsinki University Hospital, but restrictions apply to the availability of these data, which were used under license for the current study, and so are not publicly available. Data are, however, available from the authors upon reasonable request and with permission of Helsinki University Hospital.*

Articles from Acta Radiologica Open are provided here courtesy of SAGE Publications

RESOURCES