In vivo intervertebral disc deformation: intratissue strain patterns within adjacent discs during flexion–extension

Abstract

The biomechanical function of the intervertebral disc (IVD) is a critical indicator of tissue health and pathology. The mechanical responses (displacements, strain) of the IVD to physiologic movement can be spatially complex and depend on tissue architecture, consisting of distinct compositional regions and integrity; however, IVD biomechanics are predominately uncharacterized in vivo. Here, we measured voxel-level displacement and strain patterns in adjacent IVDs in vivo by coupling magnetic resonance imaging (MRI) with cyclic motion of the cervical spine. Across adjacent disc segments, cervical flexion–extension of 10° resulted in first principal and maximum shear strains approaching 10%. Intratissue spatial analysis of the cervical IVDs, not possible with conventional techniques, revealed elevated maximum shear strains located in the posterior disc (nucleus pulposus) regions. IVD structure, based on relaxometric patterns of T2 and T1ρ images, did not correlate spatially with functional metrics of strain. Our approach enables a comprehensive IVD biomechanical analysis of voxel-level, intratissue strain patterns in adjacent discs in vivo, which are largely independent of MRI relaxometry. The spatial mapping of IVD biomechanics in vivo provides a functional assessment of adjacent IVDs in subjects, and provides foundational biomarkers for elastography, differentiation of disease state, and evaluation of treatment efficacy.

High-frame-rate analysis of knee cartilage deformation by spiral dualMRI and relaxation mapping

“…The relaxation map for T1ρ was acquired by fitting the MR intensity images to a monoexponential relaxation model. 25 Voxels that had coefficient of determination (R 2 ) less than 0.66 with respect to the fitting were removed. 26 T2 mapping parameters were as follows: TEs=13, 26, 39, 52, 65, 78 ms, TR=1290 ms, FOV=80×80 mm 2 , spatial resolution=420×420 μm 2 , slice thickness=1.7 mm, slice number=13, flip angle=180°, NA=2.…”

“…The relaxation map for T1ρ was acquired by fitting the MR intensity images to a monoexponential relaxation model. 25 Voxels that had coefficient of determination (R 2 ) less than 0.66 with respect to the fitting were removed. 26 T2 mapping parameters were as follows:…”

High frame rate deformation analysis of knee cartilage by spiraldualMRIand relaxation mapping

“…The relaxation map for T 1 ρ was acquired by fitting the MR intensity images to a monoexponential relaxation model. 30 Voxels that had coefficient of determination less than 0.66 with respect to the fitting were removed. 31 T 2 maps were collected by using a multi-echo spin echo sequence, and the parameters used were as follows: TEs = 13, 26, 39, 52, 65, 78 ms; TR = 1290 ms; FOV = 80 × 80 mm 2 ; spatial resolution = 420 × 420 μm 2 ; slice thickness = 1.7 mm; slice number = 13; flip angle = 180 T 1 ρ changes after cyclic loading were measured by calculating the T 1 ρ difference before and after loading with respect to the initial T 1 ρ value.…”

Ogden material calibration via magnetic resonance cartography, parameter sensitivity and variational system identification

“…A potential path forward in the sensitivity analysis would be to use the physical, anisotropic invariants of Criscione et al [28], but instead leveraging work functions of the deformation gradient tensor. In tandem with, for example, fibre direction fields derived from MRI as in in vivo [45] or ex vivo [33] techniques, we envision coupling of multimodal information and VSI for material versus structural [46] characterization. Another potential path forward includes enhancing the observability of all material parameters in experiments based on procedurally tuning test geometry.…”