Can Quantitative Magnetic Resonance Imaging Predict Mechanical Behavior of Human Intervertebral Disks with Different Grades of Degeneration?

Introduction The dramatic changes in morphology, composition, and structure that occur in the intervertebral disk (IVD) with aging and degeneration are accompanied by specific changes in mechanical properties of the disk material.1,2 Evaluation of these changes in the IVD hinges on the ability to objectively and noninvasively assess the IVD matrix composition and integrity. Different studies on human IVDs have correlated IVD matrix composition and integrity to the longitudinal magnetization recovery T1, the transverse magnetization decay T2, the magnetization transfer ratio (MTR), and apparent diffusion coefficient (ADC).3,4 Correlations and multiple linear regressions have been also identified between quantitative magnetic resonance imaging (qMRI) parameters, biochemical, and mechanical parameters of targeted enzyme matrix denaturation and buffer-treated bovine IVDs. To this end, qMRI analysis can be used to correlate MRI signal to the mechanical properties of NP and AF tissue in order to predict structural changes in IVDs with degeneration. The aim of the present study was to determine how quantitative MRI parameters can predict biomechanical properties in human IVDs with different grades of degeneration. Materials and Methods Experimental Groups Ten whole lumbar spine specimens, 5 disks per spine, were obtained through organ donations via Héma-Québec within 24 hours after death. Age of donors was from 32 to 77 years. The samples were vacuum sealed in plastic bags for MRI to maintain hydration. MRI Procedure The MRI examinations were carried out in a 1.5T whole-body Siemens’ Avanto system using the standard circularly polarized head coil. The samples were placed in a sagittal orientation and T1, T2, MTR, and ADC were measured as previously described.1 All disks (n = 50) were then graded from T2-weighted images according to the classification system described by Pfirrmann. Numerical analysis of quantitative MRI was performed using a custom code written in MATLAB (MathWorks, Natick, MA, USA) allowing the selection of the regions of interest (ROI) and the calculation of average signal intensities from all images. ROI were traced manually as polygonal shapes with no contact with the endplate tissues and were reproduced identically on all T1, T2, Ms/Mo ratio, and diffusion images. Mechanical Testing Procedure Confined compression tests were performed on 5-mm-diameter cylindrical plugs of tissue using a custom built axial testing machine. Material parameters (