Characterization of the mechanical properties of intervertebral discs and their relationships with quantitative magnetic resonance imaging [English translation]

Intervertebral disc pathologies are a major health problem in occidental countries and thus the use of early non-invasive diagnostic tool is necessary. The first signs of intervertebral discs (IVD) degeneration is a change in their mechanical behavior and this can be detected by quantitative magnetic resonance imaging (MRI). To assess the early mechanical changes in the IVD, it is essential to establish strong correlations between mechanical properties and MRI parameters. The hypothesis of our study is that MRI parameters are sensitive to changes in IVD mechanical properties and that each mechanical parameter will be strongly correlated to a set of MRI parameters. To investigate correlations between the two parameters, mechanical properties need to be defined; however, several reports have shown that mechanical properties vary accordingly to the protocols used principally in unconfined compression. Therefore, the first part of this project consisted in defining the most reliable and the most relevant protocol to assess the mechanical properties of the annulus fibrosus and the nucleus pulposus. To this aim, caudal bovine discs were tested and mathematical models (viscoelastic, BPVE) were suited to the experimental data. The results showed that annulus and nucleus should not be tested with the same protocol, and the best experimental data were obtained with a pre-loading and a swelling of the sample. The pre-load and the swelling represent the closer physiological conditions of the disc during the mechanical test. In the literature, mechanical properties correlate moderately to MRI parameters, therefore there is a need for studying more mechanical parameters and multi-linear regression with MRI parameters for a better understanding of the relationships between the mechanical and MRI properties of the disc. Isolated discs were digested by trypsin in an ex-vivo condition. Discs digestion was performed by three different ways with varying trypsin concentrations and time of digestion, and an in-situ group was used as a control. Then, the annulus and the nucleus for each disc were subjected to a set of mechanical tests (unconfined, confined, and direct measure of permeability) associated with the MRI and biochemical assays (proteoglycan content). The novelty of the study is that the direct permeability measurement is related to MRI acquisitions of the intervertebral discs. MRI parameters used were the relaxation times (T1, T2, T2*), the magnetization transfer ratio (MTR) and the diffusion (FA and ADC) and the biochemical assays were used to investigate the proteoglycan content of tissues. Results showed that enzymatic digestion influences the mechanical properties of the disc and lower their MRI parameters. The diffusion was strongly influenced by the digestion; thus suggesting a degradation of the matrix integrity rather than changes in the biochemical contents. Biochemical assays on the disc further confirmed a clear degradation between extreme groups (in-situ and 24h digestion) but less prominent and important within each group. The low content of biochemical changes partly explains the little effect observed on the MRI signal. Weak direct correlations exist as well as moderate multiple correlations, however, direct and multiple correlations are greater if only the extreme groups are considered. With the extreme groups, a combination of MRI parameters can predict 60% variations of radial and axial permeability of the samples. Biochemical tests performed on several samples (24 per tissue) were used to qualitatively assess the effect of enzymatic digestion on the different groups. To overcome the uncertainty of the biochemical measuring assay, histological sections were carried out to further analyse the qualitative and quantitative biochemical content of the discs. The limits of this study are that the effect of hydration and enzyme activity on the discs’ MRI and mechanical parameters were not clearly identified. Our fundamental research, on the one hand, helps to a better understand the mechanical behavior of the intervertebral disc thanks to the study of the influence of experimental protocols on the mechanical response of the disc and on the other hand our study contributes to the development of a diagnostic tool from MRI-mechanical correlations. This study should used to be compared with a similar study on the dynamic behavior of the disc with the use of new MRI sequences such as CEST (Chemical exchange saturation transfer ) or T1ρ (spin-lock magnetic resonance imaging).