Introduction. Vertebral compression fractures are the most common type of osteoporotic fracture. Though 89% of clinical fractures occur anteriorly, it is challenging to replicate these ex vivo with the underlying intervertebral discs (IVDs) present. Furthermore, the role of disc degeneration in this mechanism is poorly understood. Understanding how disc morphology alters vertebral strain distributions may lead to the utilisation of IVD metrics in fracture prediction, or inform surgical decision-making regarding instrumentation type and placement. Aim. To determine the effect of disc degeneration on the vertebral trabecular bone strain distributions in axial compression and flexion loading. Methods. Eight cadaveric thoracolumbar segments (T11-L3) were prepared (N=4 axial compression, N=4 flexion). µCT-based digital volume correlation was used to quantify trabecular strains. A bespoke loading device fixed specimens at the resultant displacement when loaded to 50N and 800N. Flexion was achieved by adding 6° wedges. Disc degeneration was quantified with Pfirrmann grading and T2 relaxation times. Results. Anterior axial strains were 80.9±39% higher than the posterior region in flexion (p<0.01), the ratio of which was correlated with T2 relaxation time (R. 2. =0.80, p<0.05). In flexion, the central-to-peripheral axial strain ratio in the endplate region was significantly higher when the underlying IVDs were non-degenerated relative to degenerated (+38.1±12%, p<0.05). No significant differences were observed in axial compression. Conclusion. Disc degeneration is a stronger determinant of the trabecular strain distribution when flexion is applied. Load transfer through non-degenerate IVDs under flexion appears to be more centralised, suggesting that disc degeneration predisposes flexion-type compression fractures by shifting high strains anteriorly. Conflicts of interest. The authors declare none. Sources of funding. This work was funded by the Engineering & Physical Sciences Research Council (EP/V029452/1), and Back-to-Back

Background. In vivo evaluation of IVD strains is crucial to better understand normal and pathological IVD mechanics, and to evaluate the effectiveness of treatments. This study aimed to 1) develop a novel in vivo technique based on 3T MRI and digital volume correlation (DVC) to measure strains within IVDs and 2) to use this technique to resolve 3D strains within IVDs of healthy volunteers during extension. Methods. This study included 40 lumbar IVDs from eight healthy subjects. The optimal MR sequence to minimise DVC uncertainties was identified by scanning one subject with four different sequences: CISS, T1VIBE, T2SPACE, and T2TSE. To assess the repeatability of the strain measurements in spines with different anatomical and morphological variations four subjects were scanned with the optimal sequence, and uncertainties of the strain measurements were quantified. Additionally, to calculate 3D strains during extension, MRIs were acquired from six subjects in both the neutral position and after full extension. Results. Measurement errors were lowest when using the T2TSE sequence (precision=0.33 ± 0.10%, accuracy=0.48 ± 0.11%). The largest average maximum tensile and shear strains were seen at the L2-L3 level in all volunteers (7.2 ± 1.5% and 6.8 ± 1.1%, respectively), while the L5-S1 level experienced the lowest average tensile and shear strains (3.5 ± 1.0% and 3.9 ± 0.7%, respectively). Conclusion. The findings of this study establish clinical MRI-based DVC (MRI-DVC) as a new tool for in vivo strain measurement within human IVDs. MRI-DVC successfully provided internal strain distributions within IVDs and has great potential to be used for a wide range of clinical applications. Conflict of interest: No conflicts of interest. Source of funding: This work was supported by the EPSRC, New Investigator Award, EP/V029452/1