It is widely understood that the nucleus pulposus (NP) in non-degenerate intervertebral discs (IVDs) provides a hydrostatic pressure that causes outward bulging in the surrounding anulus fibrosus (AF).1–3 Although much of the evidence for this has been obtained through experiments conducted at low strain rates, the IVD is often subjected to dynamic loads for example in sport, and injurious events such as falls, vehicle accidents, airplane ejections, and blast-related incidents.4,5 The mechanical role of the NP has not been investigated at loading rates relevant to these loading conditions.
At low strain rates, Meakin et al6 compressed sheep vertebral body-disc-vertebral body (VB-disc-VB) samples that had been sliced along the sagittal plane and pushed up against a perspex sheet, with the NP present, and also once the NP had been removed. Before removal of the NP the inner boundaries of the AF moved outwards when compressed, however, after removal of the NP, the inner AF moved inwards, suggesting that the NP plays a role in internally pressuring the IVD such that the AF inner boundary is forced outwards, rather than inwards and thus reducing the shear stresses within the AF. Seroussi et al7 also found inward bulging of the inner wall of the AF in compression and flexion after removing the NP in human motion segments. They embedded 0.5 mm diameter beads within the IVD, which, through obtaining radiographs during the loading, allowed analysis of the internal motion.
Markolf and Morris8 compressed human VB-disc-VB samples intact, but also with the endplates directly above and below the NP, and the NP itself, removed. They found, at quasi-static rates (∼0.04 mm/s), similar force–displacement responses between the isolated AF and that of the intact disc, suggesting that the relative importance of the internal pressure provided by the NP was small in comparison to the role of the AF in influencing the behavior of the disc under compression.
The mechanical function of the NP when the IVD is loaded in daily activities, sport, and during injury is unclear. While some studies have shown that removal of the NP affects the behavior of the inner AF,7,9 others have demonstrated that at low loading rates removal of the NP pressure does not affect the stiffness of the IVD.8 No studies have investigated the effect of the NP pressure on the behavior of the IVD when exposed to high-rate external loading.
We hypothesize that removal of the pressure in the NP will influence the stiffness and energy-absorbing capacity of the IVD because the NP pressure has been shown to be critical in ensuring that the inner AF bulges outwards under compression. The aim of this study was to investigate the effect of depressurising the NP on the force–displacement response, and the amount of energy absorption in IVDs when compressed at high strain rates.
MATERIALS AND METHODS
Eight bovine tails were obtained from a local butcher and Computed Tomography (CT) scanned (Siemens AS64, Erlangen, Germany—slice thickness 1 mm, pixel size, 0.4 × 0.4 mm) to check for vertebral fractures or any other signs of pathology. Specimens were stored frozen at –20 °C and thawed overnight at room temperature before dissection and testing. Two separate motion segments were obtained from each tail by cutting transversely through the first, second, and third caudal VB at mid-height resulting in 16 motion segments for testing. Throughout the preparation process the samples were regularly sprayed with phosphate buffered saline (0.15 m/L) to keep them hydrated.
Using a custom-built alignment jig the superior VB of the segment was positioned such that the mid-plane of the disc was parallel to the ends of, and centered within, a 90 mm diameter pot. The superior VB was then fixed in position using polymethyl-methacrylate (PMMA) bone cement before being turned upside down, allowing the inferior VB to be lowered into a second pot and again secured into position using PMMA.
Of the 16 samples, eight were used for high rate testing, and eight to quantify the change in NP pressure caused by drilling through the center of the endplate into the disc.
Effect of Drilling Through the Endplate on the NP Pressure
Each of the eight intact samples were placed within a uniaxial testing machine (Instron 5866, High Wycombe, UK), and compressed to 50 N. The displacement at this load was recorded and the cross-head was held steady for the remainder of this part of the experiment. A 1.3 mm diameter pressure needle (8CT/4F/SS/HP, Gaeltec Ltd, UK) was used to measure the pressure profile across the IVD. To allow the needle to be inserted through the center of the IVD a 20 G guide needle (0.9 mm diameter) was used to form the path for the wider pressure needle. A calibrated potentiometer (Caldaro, 18FLPA50, Stockholm, Sweden) was attached to the pressure needle using a custom-made fixation rig and touched up against the side of the IVD, such that displacement could be measured as the pressure needle was pulled through the IVD. The pressure transducer was aligned such that the diaphragm faced in the cranial direction and the pressure profile measurements were repeated three times. Data were recorded at 1000 Hz (cDAQ-9174, National Instruments, Newbury, UK).
Following the intact pressure profile measurements, a 5.5 mm hole was drilled through the cranial VB and endplate into the NP. Care was taken to ensure that the hole was deep enough that it had passed through the cranial endplate, but had not disrupted the caudal endplate. This was confirmed using a fluoroscope image (Figure 1) (InsightFD Mini-C-Arm, Fluoroscan, MA). The sample was then placed back within the uniaxial testing machine, compressed to 50 N, and held at that position. The cross-head displacement was recorded to capture the change in disc height before and after drilling the hole. The new pressure profile was measured using the same hole used for the intact measurements.
The NP pressures of the intact, and drilled samples were compared at the center of the disc, which was located by identifying the midpoint between the entry and exit points of the needle. The entry and exit points were defined as the positions where the pressure increased above, or dropped below, 20 kPa. The pressure in the NP was taken as an average of the measurements 1 mm either side of the midpoint. Statistical differences were assessed using paired t tests with the significance level set at P = 0.05.
Drop Rig Tests
Drop rig tests were carried out on the second set of eight samples. The pressure transducer was not used in these samples to ensure that its insertion into the disc would not affect the observed response under load. The experiments were carried out using a custom-built drop-weight rig, which allowed a 2.5 kg mass to be dropped onto each sample from a range of heights (Figure 2). The motion segments were positioned centrally under a 50 mm diameter tup attached to the falling mass and above a 6-axis, dynamic load cell (M3944, Sunrise Instruments Ltd, MI). Uniaxial accelerometers (352C04, PCB-Piezotronics, NY) were fixed to the falling mass and to both the superior and inferior mounting pot. All data were recorded using a PXIe system at 25 kHz (1082, National Instruments, Newbury, UK). Data were filtered using a low-pass Butterworth filter with a cut off frequency of 1 kHz. High speed video (PhantomV12.1, Vision Research, Bedford, UK) was captured at a rate of 12,500 Hz.
For the intact tests the segment was subjected to impacts from a range of heights up to 64 cm. A maximum height of 64 cm was chosen since preliminary experiments had shown that samples were likely to be damaged if impacted from heights greater than this. Heights were incrementally doubled starting from 2 cm. Between drops, the samples were allowed to relax for 15 minutes. Preliminary investigations showed that this time period was sufficient to obtain a repeatable force–displacement response from the same drop test (<10% differences in peak force and time-to-peak). The force–time curve from the 2 cm drop height was used as a baseline result. After every increase in drop height, a test from 2 cm was repeated to ensure that the segment had not failed and the results were repeatable (<10% difference in peak force between tests). Post-impact CT scans were taken of the samples and if fractures were seen in the VBs the results from this sample were removed from the analysis.
As with the pressure profile tests, following the intact test, a 5.5 mm hole was drilled through the cranial VB and endplate into the NP. The same test protocol that was used for the intact segments was then repeated on the drilled NP segments. The energy absorbed by each IVD was calculated, assuming it being equal to the loss in kinetic energy of the impactor; this was calculated for each test from the initial and rebound velocities, determined using the high-speed video footage. Differences in peak forces, peak displacements, strain rates, and percentage energy absorbed between the intact and drilled NP samples were assessed using paired t tests with the significance set to P = 0.05.
Effect of Drilling Through the Endplate on the NP Pressure
On average, the discs reduced in height by 0.50 ± 0.43 mm (average ± one standard deviation) after the 5.5 mm diameter hole was drilled through the endplate into the disc; this corresponds to a 6.8 ± 5.5% reduction in central disc height (P = 0.013). The NP pressure was 28.5 ± 15.5% lower after the 5.5 mm hole was drilled through the endplate into the NP (Figure 3, P = 0.004).
Drop Rig Tests
Post-impact CT scans revealed fractures in the VBs of one of the segments and therefore results from this test were not included in the analysis. No fractures were seen in the post-impact CT scans of any other segment and the force–time responses of the drops from 2 cm were consistent when the segment was both intact and with a depressurized NP (the peaks were within 10%). Therefore, it was assumed that none of these segments had failed during testing. Force–displacement responses for each sample, intact and with a depressurized NP, are shown in Figure 4.
The average strain rate in reciprocal seconds (/s) from the 2, 4, 8, 16, 32, and 64 cm drops were 89 ± 30, 142 ± 34, 237 ± 34, 243 ± 49, 291 ± 66, and 534 ± 131, respectively. No significant differences were seen in the strain rates for intact and drilled samples (average P value from paired t tests at each drop height = 0.42 ± 0.25). The average peak force and peak displacement for each drop height are shown in Figure 5. No significant differences were seen between the intact and drilled NP groups at any drop height (average P value from paired t tests at each drop height = 0.32 ± 0.25 for the peak force, and 0.82 ± 0.16 for the peak displacement).
On average, the intact IVDs absorbed 71.8 ± 6.6% of the input energy (Figure 6). There was no rate dependency in the percentage energy absorbed by the IVDs, nor was there any difference in the percentage energy absorbed in the intact and depressurized NP samples (average P value from paired t tests at each drop height = 0.26 ± 0.14).
The motivation of this work was to investigate the mechanical role of the NP in IVDs when compressed at high strain rates. The hypothesis that reducing the NP pressure would affect the stiffness, and energy-absorbing capacity of the IVD was disproved since no statistical difference was seen in the response of the intact IVDs and those that had been depressurized by 28.5%. This demonstrates that the pressure provided by the NP does not affect the transfer of load through, or energy absorbed by, the IVD at high rates of compressive loading. It can therefore be inferred that the AF, rather than the NP, plays the most important role in transferring load, and absorbing energy in the IVD at high strain rates.
To avoid the complicating factor of IVD degeneration, which is common in human cadaveric samples, bovine IVDs were used for this study. The mechanical behavior,10 swelling pressure,11 composition,12 and relative cross-sectional areas and height13 (and therefore volume) of bovine IVDs have been found to be similar to human samples, and have therefore been suggested as a good model for studies of this kind.14 However, there are geometric differences between bovine and human IVDs; human endplates are flatter, while bovine discs have convex endplates; and bovine tail samples do not have posterior elements. These differences, along with the variations in physiological loading to which tail discs are exposed in comparison to the human spine,14 mean that future investigations to confirm the results found here in human samples are recommended. The percentage energy absorbed calculated in this study may have been affected by both friction between the falling mass and the guide bars of the drop rig, and compliance in other components of the experimental setup. Therefore, the values found here likely exaggerate the energy absorption characteristics of the IVDs, however comparisons of rate effects, and the drilled and intact specimens are reasonable. Additionally, we have only studied one mode of loading here; pure axial compression. Physiologically, human IVDs are exposed to other modes of loading which also warrant investigation, particularly because it has been shown that NP removal causes an amplification of deformations in other modes at lower rates of loading.15–17
Although previous studies have shown that decompressing the NP causes greater peak stresses in the posterior AF,18 thus suggesting a shift in load bearing from the NP to the AF, our study did not find any difference in disc stiffness following removal of NP pressure. This is similar to the findings of Markolf and Morris,8 who concluded that the NP plays a less important role than the AF in determining the stiffness of the IVD following low rate axial compression tests on human cadaveric samples. Other studies that have compared internal deformation of intact and denucleated IVDs have found that the inner AF boundary bulged inwards following NP removal.6,7 It was not possible, however, to analyze the internal deformations of the IVDs in this study. Therefore, it is not known whether the AF bulged inwards or outwards following depressurization of the NP at these high rates of loading. Regardless, if there was a change in the direction of deformation of the inner wall of the AF, the AF appears to be the component of the IVD that dominates the contribution to the stiffness, and energy-absorbing capacity of the IVD.
The previous understanding of the function of the NP was that it provides a hydrostatic pressure that forces the AF fibres into tension such that they are able to resist compressive forces.1–3,19 Had the NP pressure been having any mechanical function during the loading of the intact motion segments in this study, a significant difference would have been seen in the behavior of the IVD following reduction of the NP pressure. Although the NP pressure may have a mechanical role in other modes of loading, the fact that no significant differences were seen in either the stiffness or the energy-absorbing capacity of the IVD means that in pure axial compression, and at high rates of loading, the NP does not have a mechanical function. It is likely that the mechanical role of the NP is in fact to restore, and maintain the height of the disc, rather than to carry any load. The results of this study also suggest that loads transfer through the VB cortical bone to the peripheral AF, and not through the bone proximal to the endplate that was disrupted by the 5.5 mm hole.
These results have implications for computational modellers aiming to replicate the axial compressive behavior of IVDs. It is common to define the pressure within the NP or assign material properties to the NP in computational models. The results from this study suggest that the NP material properties and pressure are not important in comparison to the material properties of the AF when replicating the mechanical behavior of the IVD. Therefore, future research efforts should prioritize obtaining accurate material models of components of the AF, rather than the NP in order to improve the accuracy of IVD models.
Clinically, disc decompression is common in the thoracic and lumbar spine.20 The results presented in this study suggest that surgical reconstruction for the NP may not be necessary for restoring the axial compressive stiffness, and energy-absorbing capacity of the IVD at high rates of loading. Rather, restoring the mechanical function of the AF, coupled with the low strain rate behavior of the NP may be of higher importance. These results are of interest to developers of NP-replacement devices as it indicates that efforts should concentrate on ensuring that the devices restore and maintain IVD height, so that loading of the zygapophysial joints is not increased,21 rather than concentrating on restoring the stiffness and energy-absorbing characteristics of the NP.
1. Keyes DC, Compere EL. The normal and pathological physiology of the nucleus pulposus
of the intervertebral disc
. J Bone Jt Surg Am
2. Nachemson A. Lumbar intradiscal pressure: experimental studies on post-mortem material. Acta Orthop
3. Nachemson A, Elfström G. Intravital dynamic pressure measurements in lumbar discs. A study of common movements, maneuvers and exercises. Scand J Rehabil Med Suppl
4. Panzer M, Fice J, Cronin D. Cervical spine
response in frontal crash. Med Eng Phys
5. Yoganandan N, Ray G, Pintar FA, et al. Stiffness and strain energy criteria to evaluate the threshold of injury to an intervertebral joint. J Biomech
6. Meakin JR, Reid JE, Hukins DW. Replacing the nucleus pulposus
of the intervertebral disc
. Clin Biomech (Bristol, Avon)
7. Seroussi RE, Krag MH, Muller DL, et al. Internal deformations of intact and denucleated human lumbar discs subjected to compression, flexion, and extension loads. J Orthop Res
8. Markolf KKL, Morris JJM. The structural components of the intervertebral disc
. J Bone Jt Surg
9. Meakin JR, Redpath TW, Hukins DW. The effect of partial removal of the nucleus pulposus
from the intervertebral disc
on the response of the human annulus fibrosus to compression. Clin Biomech (Bristol, Avon)
10. Beckstein JC, Sen S, Schaer TP, et al. Comparison of animal discs used in disc research to human lumbar disc: axial compression mechanics and glycosaminoglycan content. Spine (Phila Pa 1976)
11. Oshima H, Ishihara H, Urban J, et al. The use of coccygeal discs to study intervertebral disc
metabolism. J Orthop Res
12. Demers CN, Antoniou J, Mwale F. Value and limitations of using the bovine tail as a model for the human lumbar spine
. Spine (Phila Pa 1976)
13. O’Connell GD, Vresilovic EJ, Elliott DM. Comparison of animals used in disc research to human lumbar disc geometry. Spine (Phila Pa 1976)
14. Alini M, Eisenstein SM, Ito K, et al. Are animal models useful for studying human disc disorders/degeneration? Eur Spine J
15. Meakin JRR, Hukins DWLW. Effect of removing the nucleus pulposus
on the deformation of the annulus fibrosus during compression of the intervertebral disc
. J Biomech
16. Iatridis JC, MacLean JJ, Ryan DA. Mechanical damage to the intervertebral disc
annulus fibrosus subjected to tensile loading. J Biomech
17. Heuer F, Schmidt H, Claes L, et al. A new laser scanning technique for imaging intervertebral disc
displacement and its application to modeling nucleotomy. Clin Biomech
18. Dolan P, Luo J, Pollintine P, et al. Intervertebral disc
decompression following endplate damage: Implications for disc degeneration depend on spinal level and age. Spine (Phila Pa 1976)
19. Newell N, Little J, Christou A, et al. Biomechanics of the human intervertebral disc
: a review of testing techniques and results. J Mech Behav Biomed Mater
20. Crock HV. Internal disc disruption. A challenge to disc prolapse fifty years on. Spine (Phila Pa 1976)
21. Dunlop RB, Adams MA, Hutton WC. Disc space narrowing and the lumbar facet joints. J Bone Joint Surg Br