The Effect of Degeneration on Internal Strains and the Mechanism of Failure in Human Intervertebral Discs Analyzed Using Digital Volume Correlation (DVC) and Ultra-High Field MRI

Abstract

The intervertebral disc (IVD) plays a main role in absorbing and transmitting loads within the spinal column. Degeneration alters the structural integrity of the IVDs and causes pain, especially in the lumbar region. The objective of this study was to investigate non-invasively the effect of degeneration on human 3D lumbar IVD strains (n = 8) and the mechanism of spinal failure (n = 10) under pure axial compression using digital volume correlation (DVC) and 9.4 Tesla magnetic resonance imaging (MRI). Degenerate IVDs had higher (p < 0.05) axial strains (58% higher), maximum 3D compressive strains (43% higher), and maximum 3D shear strains (41% higher), in comparison to the non-degenerate IVDs, particularly in the lateral and posterior annulus. In both degenerate and non-degenerate IVDs, peak tensile and shear strains were observed close to the endplates. Inward bulging of the inner annulus was observed in all degenerate IVDs causing an increase in the AF compressive, tensile, and shear strains at the site of inward bulge, which may predispose it to circumferential tears (delamination). The endplate is the spine's "weak link" in pure axial compression, and the mechanism of human vertebral fracture is associated with disc degeneration. In non-degenerate IVDs the locations of failure were close to the endplate centroid, whereas in degenerate IVDs they were in peripheral regions. These findings advance the state of knowledge on mechanical changes during degeneration of the IVD, which help reduce the risk of injury, optimize treatments, and improve spinal implant designs. Additionally, these new data can be used to validate computational models.

Keywords: digital volume correlation (DVC); disc degeneration; endplate fracture; internal 3D strain; intervertebral disc (IVD); magnetic resonace imaging (MRI); vertebral fracture.

Figure 1

Schematic of (A) the MRI compatible loading rig within an external custom-made screw-driven compression rig used to load samples to 50 N and 1 kN, and (B) the MRI compatible loading rig within the materials testing machine used for failure tests. The applied loads were purely axial, with no bending or shear movements. A mid-coronal MRI slice of a representative degenerate sample is shown unloaded in (C), loaded to 1kN in (D), and after failure in (E). The endplate fracture location is denoted by an asterisk.

Figure 2

The endplate centroid was located using a method previously developed by Little et al. (2007). Four lines were drawn around the mid-transverse plane of the IVD to create a rectangle: line 1 passed through the two inflections of the posterior annulus boundary (points 1 and 2), lines 2 and 3 were perpendicular to line 1 and passed through the two most lateral points of the annulus boundary (points 3 and 4), and line 4 was perpendicular to lines 2 and 3 passing through the most anterior point of the annulus boundary (point 5). The geometric center of this rectangle was then used as the endplate centroid. The percentages refer to the distance between the furthest of either the most anterior (point 5) or most posterior (points 1 or 2) locations of the IVD in the mid-transverse plane.

Figure 3

Axial photographs of typical non-degenerate (left, Pfirrmann grade = 2, 26-year-old) and degenerate (right, Pfirrmann grade = 4, 53-year-old) samples taken during specimen dissection after the failure test.

Figure 4

(A) Maximum 3D principal (tensile) strain, and (B) maximum 3D shear strain maps are shown for a typical sample on the same coronal slice. Black arrows show the location of peak strains close to the endplate.

Figure 5

Bar plots of the 3D strain distributions in non-degenerate (ND) and degenerate (D) IVDs in the four anatomical regions (LAF = lateral annulus fibrosus, NP = nucleus pulposus, AAF = anterior annulus fibrosus, and PAF = posterior annulus fibrosus) for the (A) axial (ε_yy), (B) minimum principal, (C) maximum principal, and, (D) maximum 3D shear strain components (left). The bars, and error bars represent the average and SD of strain values, respectively and an asterisk (*) denotes significant differences (p < 0.05). Transparent views of 3D strain maps for a typical non-degenerate and degenerate IVD are shown on the right.

Figure 6

Comparisons of strains seen in specimens with (A) outward bulging of the inner AF (Group OB) and (B) inward bulging of the inner AF (Group IB). (A,B) show schematics of the outward and inward bulge of the inner AF, respectively. (i) shows the mid-coronal MRI slice of typical specimens in each group, with (ii) showing the left and right lateral regions with yellow dashed lines delineating the morphology of lamellae. The MRI slice, with superimposed minimum 3D principal strains and maximum 3D shear strains are shown in (iii) and (iv), respectively. (C) compares the 3D principal and 3D shear strains in the two groups and (D) compares the axial and minimum principal strains in the two groups. Error bars represent standard deviations.

Figure 7

Representative fluoroscope (A,B) and MR (C,D) images of a sample before (A,C) and after (B,D) fracture. Combining these two imaging modalities, along with manual dissection of each sample allowed the location of fracture to be identified, with the fluoroscope images being used to identify the initiation of the fracture and the MR images being used to precisely locate the fracture site on a transverse slice. Yellow arrows identify the location of fracture.

Figure 8

(A) Normalized locations of endplate fracture in each of the samples tested. Squares (green) are non-degenerate samples, circles are degenerate (red). Solid points are cranial, and hollow points represent caudal endplate fracture. Concentric rings represent zones 20, 40, 60, and 80% of the distance between the furthest of either the most anterior or the most superior points of the vertebral body in the sagittal plane. (B–F) are axial strain distributions in the transverse plane of the degenerate IVDs overlaid by the location of endplate fracture. Non-degenerate discs are not shown as there was no correlation between regions of high axial strains and fracture location. Regions inside of the dashed lines indicate the high axial strain regions, and the areas outside represent low axial strain regions. Black crosses denote locations of fracture in the strain maps.