Introduction: Nutrients that nourish cells in the avascular disc diffuse from the vertebral capillaries, across the cartilage endplate (CEP), and into the nucleus pulposus (NP). Thus, the composition of CEP tissue and subchondral bone marrow may play important roles in disc health by influencing nutrient diffusion and vertebral perfusion. However, the relative roles of these factors in disc degeneration have not been quantified. The goal of this study was to use biomarkers of tissue composition derived from quantitative MRI to establish how CEP composition (surrogate for permeability) and vertebral bone marrow fat fraction (BMFF, surrogate for perfusion) contribute to disc degeneration in humans. This knowledge could help clarify the etiology of disc degeneration and identify factors impacting the efficacy of regenerative intradiscal biologic therapies.

Methods: MRI data from 77 patients were included in this study. Sixty subjects had chronic low back pain (cLBP; >3 month duration; age=40.0±11.9 years, 19–65 [mean±SD, min–max]; 28 female, 32 male; ODI=26.1±14.6; VAS=6.5±2.4,) and 17 were asymptomatic controls (age=41.4±10.9 years, 23–67; 9 female, 8 male). Ultra-short echo-time (UTE) MRI was used to calculate CEP T2* relaxation times (reflecting biochemical composition and permeability), chemical shift encoding-based water-fat MRI was used to calculate vertebral BMFF (proportional to the amount of hematopoietic marrow and providing a proxy for perfusion), and T1ρ MRI was used to calculate T1ρ relaxation times in the NP (NP T1ρ, reflecting proteoglycan content and degenerative grade). A combination of neural network-generated and manually-generated segmentations for levels L4–S1 were used for biomarker analysis (Fig. 1). Univariate linear regression was used to assess the independent effects of CEP T2* and vertebral BMFF on NP T1ρ. Mixed effects linear regression accounting for spinal level, age, and sex were used to assess combined relationships.

Results: CEP T2* and vertebral BMFF were independently associated with NP T1ρ (p=0.014 and 0.0001, respectively). After adjusting for age, sex, and spinal level, NP T1ρ remained significantly associated with CEP T2* (p=0.004, Table 1) but not vertebral BMFF (p=0.40). The relationship between disc degeneration and CEP composition was similar in cLBP and asymptomatic cohorts (p=0.59 for the interaction between cohort and CEP T2*). In this age-adjusted model, the severity of disc degeneration was also similar between cohorts (mean NP T1ρ=64.2 ms [95% CI: 59.7, 68.6] versus 71.8 ms [63.8, 79.7], cLBP versus control, p=0.09).

Discussion: These results provide the first evidence that lumbar disc degeneration severity associates with compositional deficits in the CEP. We found that CEPs with lower T2* values—reflecting lower CEP hydration, lower GAG content, a higher ratio of collagen-to-GAG, and lower permeability—were associated with more severe disc degeneration. Unlike vertebral BMFF, CEP composition was a significant predictor of disc health after accounting for co-variates, demonstrating the relative importance of CEP composition in disc degeneration in humans both with and without cLBP and with and without deficits in vertebral perfusion. The UTE-derived CEP T2* biomarker may be a promising diagnostic measurement to help select patients with adequate CEP composition to support intradiscal biologic therapies, which inherently increase nutrient demands.