Mechanical Deformation and Glycosaminoglycan Content ...cneu/NeuLab/Publications_files/ChanDD_etal_2011b.pdfDegeneration Model Eight skeletally mature, male New Zealand white rabbits

BASIC SCIENCE

1438 www.spinejournal.com August 2011

SPINE Volume 36, Number 18, pp 1438–1445©2011, Lippincott Williams & Wilkins

Copyright © 2011 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

Mechanical Deformation and Glycosaminoglycan Content Changes in a Rabbit Annular Puncture Disc Degeneration Model

Deva D. Chan , MS , * Safdar N. Khan , MD , † Xiaojing Ye , MD, PhD , † Shane B. Curtiss , AS , † Munish C. Gupta , MD , † Eric O. Klineberg , MD , † and Corey P. Neu , PhD , *

Study Design. Evaluation of degenerated intervertebral discs from a rabbit annular puncture model by using specialized magnetic resonance imaging (MRI) techniques, including displacement encoding with stimulated echoes and a fast-spin echo (DENSE-FSE) acquisition and delayed gadolinium-enhanced MRI of cartilage (dGEMRIC). Objective. To evaluate a rabbit disc degeneration model by using various MRI techniques. To determine the displacements and strains, spin-lattice relaxation time (T 1 ), and glycosaminoglycan (GAG) distribution of degenerated discs as compared to normal and adjacent level discs. Summary of Background Data. Annular puncture of the intervertebral disc produces disc degeneration in rabbits. DENSE-FSE has been previously demonstrated in articular cartilage for the measurement of soft tissue displacements and strains. MRI also can measure the T 1 of tissue, and dGEMRIC can quantify GAG concentration in cartilage. Methods. In eight New Zealand white rabbits, the annulus fi brosis of a lumbar disc was punctured. After 4 weeks, the punctured and cranially adjacent motion segments were isolated for MRI and histology. MRI was used to estimate the disc volume and map T 1 . DENSE-FSE was used to determine displacements for the estimation of strains. dGEMRIC was then used to determine GAG distributions. Results. Histology and standard MRI indicated degeneration in punctured discs. Disc volume increased signifi cantly at 4 weeks after the puncture. Displacement of the nucleus pulposus was distinct

Low back pain, which affects more than 25 million peo-ple in the United States, has serious personal and soci-etal impact. 1 Although the etiology of this pain remains

largely unknown, there is a clinical link between degenerate intervertebral discs and chronic low back pain. 2 Disc degen-eration is an age-related process that involves both biochemi-cal and mechanical changes. 3

Animal models of intervertebral disc degeneration can enhance understanding of the progression of disc degenera-tion and provide a means to evaluate treatment regimens. 4 The rabbit annular puncture model 5 , 6 can be evaluated by using various techniques. Histology allows for the localiza-tion of microscale disc structure. Mechanical testing of the disc throughout the degeneration process in an animal model provides valuable information about the functional integrity of the disc. Despite the utility of these techniques, sacrifi ce of the animal is required, and this limits the number of speci-mens that can be examined at each time point in longitudinal studies. Instead, noninvasive imaging techniques can be used to evaluate the changes that occur in longitudinal studies of animal models.

Noninvasive medical imaging can be used to monitor changes in animal models of disc degeneration. Although pre-vious studies of the rabbit puncture model used radiography to evaluate changes in disc height index 5 and the morphology of vertebral body bone, 6 measuring disc height from projection images can result in variations in disc height indexes. 7 Mag-netic resonance imaging (MRI), a technique more appropriate than radiography for directly visualizing soft tissues, can be

From the * Weldon School of Biomedical Engineering, Purdue University, West Lafayette, Indiana ; † Lawrence J. Ellison Musculoskeletal Research Center, Department of Orthopaedic Surgery, University of California Davis Medical Center, Sacramento, California .

Acknowledgement date: January 30, 2010. Revision date: July 13, 2010. Acceptance date: July 19, 2010.

The manuscript submitted does not contain information about medical device(s)/drug(s).

Scoliosis Research Society Small Exploratory Grant funds were received in support of this work. No benefi ts in any form have been or will be received from a commercial party related directly or indirectly to the subject of this manuscript.

Chan and Khan shared fi rst authorship

Address correspondence and reprint requests to Corey P. Neu, PhD, Weldon School of Biomedical Engineering, Purdue University, 206 S. Martin Jischke Drive, West Lafayette, IN 47907; Email: [email protected]

from that of the annulus fi brosis in most untreated discs but not in punctured discs. T 1 was signifi cantly higher and GAG concentration signifi cantly lower in punctured discs compared with untreated adjacent level discs. Conclusion. Noninvasive and quantitative MRI techniques can be used to evaluate the mechanical and biochemical changes that occur with animal models of disc degeneration. DENSE-FSE, dGEMRIC, and similar techniques have potential for evaluating the progression of disc degeneration and the effi cacy of treatments. Key words: rabbit disc degeneration , displacement-encoded MRI , dGEMRIC , cartilage elastography , quantitative MRI. Spine 2011 ; 36 : 1438 – 1445

DOI: 10.1097/BRS.0b013e3181f8be52

BRS204165.indd 1438BRS204165.indd 1438 23/07/11 11:04 AM23/07/11 11:04 AM

Spine www.spinejournal.com 1439

BASIC SCIENCE Mechanical Deformation and GAG Content Changes • Chan et al


used to grade the extent of degeneration in animal models 5 , 8 and to quantify changes in the area and signal intensity of the nucleus pulposus. 6 Studies in humans, however, have shown that disc height and signal intensity may not be sensitive to the early changes that occur in disc degeneration. 9 Although MRI is well suited to provide morphologic and structural informa-tion, some quantitative MRI techniques have shown promise for detecting changes that occur in early disc degeneration. 10 , 11

Quantitative MRI is used to characterize the mechanical and biochemical characteristics of soft tissue, including mea-sures of the displacement and strain patterns in soft tissues and glycosaminoglycan (GAG) content. Displacement encod-ing with stimulated echoes and a fast-spin echo (DENSE-FSE) readout is a phase-contrast MRI technique that allows for the direct measurement of displacements that occur during cyclic loading. DENSE-FSE has previously been used in articular cartilage explants 12 and intact tibiofemoral joints. 13 Delayed gadolinium-enhanced MRI of cartilage (dGEMRIC) is a quantitative MRI technique that determines the fi xed charge density and GAG distribution within tissue by using the contrast agent gadopentetate dimeglumine (Gd-DTPA 2 − ). 14 Gadolinium-enhanced imaging has previously been used to improve evaluation of intervertebral discs 15 and to quantify GAG in the nucleus pulposus. 16 Rabbit model disc biochem-istry has been quantifi ed with high-fi eld MRI, 17 but dGEM-RIC has not been used to evaluate GAG content in an animal model of disc degeneration. Finally, despite the importance of biomechanics in disc function, displacement-encoded MRI and elastography have not yet been applied to either healthy or degenerated intervertebral discs.

To ultimately address repair and regeneration of interverte-bral disc degeneration in longitudinal studies, the purpose of this article is to evaluate a rabbit annular puncture disc degen-eration model by using quantitative MRI. The objectives were threefold, namely, to (1) evaluate the morphologic changes in the annular puncture model of disc degeneration by histology and MRI, (2) determine the displacement and strain patterns by DENSE-FSE, and (3) quantify changes in GAG concentra-tion by dGEMRIC.

MATERIALS AND METHODS

Degeneration Model Eight skeletally mature, male New Zealand white rabbits (weight ∼ 3.5 kg) were treated in an annular puncture degen-eration model, 5 while three were used as controls, under Insti-tutional Animal Care and Use Committee approval. For each treated rabbit, the intervertebral disc between the fourth and fi fth vertebrae (L4–L5) was punctured by using a 16-gauge needle under aseptic conditions. Lateral radiographs of the lumbar spine were taken during surgery and before sacrifi ce. Rabbits were killed 4 weeks postoperation, and the lumbar spines, which include the cranial and untreated disc (L3–L4) and L4–L5 discs, were isolated. Lumbar spines were also iso-lated from the control animals.

L3–L4 and L4–L5 motion segments were trimmed to in-clude the disc and approximately 5 mm of the cranial and

caudal vertebral bodies. Specimens were mounted in an MRI-compatible, cyclic loading device with cyanoacrylate and bathed in phosphate-buffered saline (PBS). 12

Magnetic Resonance Imaging All motion segments were imaged by using a vertical 400 MHz MRI system (9.4 Tesla, Bruker Medical GMBH, Ettlingen, Germany). Fast-spin echo (FSE) scans were acquired in ad-jacent coronal slices through the disc by using the following parameters: repetition time (TR) = 3000 milliseconds, echo time (TE) = 5.3 milliseconds, fi eld of view (FOV) = 25.6 × 25.6 mm 2 , matrix size = 256 × 256 pixels, slice thickness = 1.0 mm, interslice distance = 1.0 mm, and number of slices = 9. The adjacent 1-mm slices were used to estimate the in-tervertebral disc volume with ImageJ (National Institutes of Health, Bethesda, MA). Regions of interest (ROIs) containing the intervertebral disc were manually segmented in all image slices that passed through the disc. The ROI areas were then used to estimate disc volume. The initial FSE scans were addi-tionally used to select a coronal slice across the largest width of the disc for subsequent imaging. After slice selection, speci-mens were not repositioned or removed until the completion of all imaging.

Displacement Encoding with Stimulated Echoes and Fast-Spin Echo Displacement and strain patterns were determined by DENSE-FSE. A computer-controlled electropneumatic system was used to synchronize cyclic compression and DENSE-FSE imaging. 12 Every 3.0 seconds, 30 N of compression was ap-plied for 1.5 seconds, with MRI readout during this loading plateau. To prevent blurring of images due to changes in the deformation of the disc between cycles, DENSE-FSE images were not acquired until after 200 cycles of compression to ensure a steady-state load-displacement response. 18 A displacement-encoding gradient was applied before and during loading. For calculation of displacements, DENSE-FSE requires phase information from a reference scan, during which the displacement-encoding gradient off, and scans with displacement-encoding gradients in orthogonal directions (i.e. x, y). 12 Imaging parameters were as follows: effective TR = 3000 milliseconds; effective TE = 24.7 milliseconds; interleave factor = 8; FOV = 16 × 16 mm 2 ; matrix size = 128 × 128 pixels; slice thickness = 1.0 mm; and number of averages = 4. Displacement-encoding gradients in the load-ing and transverse directions were 95 mT/m, with an effective duration of 1.0 milliseconds. Using a standard FSE sequence, images of the deformed specimen were also acquired, and the deformed disc was manually segmented by using Paravision (Bruker Medical GMBH, Ettlingen, Germany).

Image processing software (Matlab v7.0, The Mathworks, Natick, MA) was used to determine displacements and es-timate Green-Lagrange strains from smoothed displacement fi elds. 12 For comparisons of trends between discs, displace-ments and strains along vertical lines of interest, which were chosen along the geometric center of the full ROI, were normalized along the disc depth.





Delayed Gadolinium-Enhanced Magnetic Resonance Imaging of Cartilage GAG patterns were determined by dGEMRIC, using MR ac-quisitions before and after the administration of an anionic contrast agent. 14 A variable TR multislice multiecho (MSME) scan was acquired through the selected slice with the follow-ing parameters: TRs = 100, 200, 300, 500, 900, and 4000 milliseconds; TE = 7.2 milliseconds; FOV = 16 × 16 mm 2 ; matrix size = 256 × 256 pixels; and slice thickness = 1.0 mm. The PBS bath was then replaced with PBS containing 2 mmol/L Gd-DTPA 2 − . Suffi cient time ( > 9 hours) was allowed for the gadolinium to penetrate into the immersed disc. A set of MSME scans was then acquired under the same imaging parameters, except that TRs = 100, 200, 300, 500, 900, and 2500 milliseconds.

MR imaging software (Paravision 3.0, Bruker Medical GMBH, Ettlingen, Germany) was used to calculate the spin-lattice relaxation time constants (T 1 ) within the imaged slice before and after the administration of Gd-DTPA 2 − . An ROI in the MSME image of the undeformed disc was then defi ned manually by using closed Bezier curves. Computational soft-ware (Matlab) was then used to calculate fi xed charge density and GAG concentration. T 1 , before and after Gd-DTPA 2 − administration, and GAG concentration, were averaged across the disc.

Histology All motion segments were fi xed in a 10% formaldehyde solu-tion immediately after the completion of all MRI tests. Fixed specimens were decalcifi ed by using 10% formic acid and im-bedded in paraffi n before slicing. Histologic slices were then stained with hematoxylin and eosin and visualized by light microscopy.

Statistics T 1 values and GAG content between control L3–L4 and con-trol L4–L5 and between untreated L3–L4 and punctured L4–L5 were compared by using a paired Student t test. P < 0.05 indicated signifi cance. All values were reported as mean ± standard error of the mean.

RESULTS

Degeneration Model Morphology Morphologic changes were observed in degenerated discs ( Figure 1 ). Lateral radiographs ( Figure 1 A) confi rmed that the discs between the fourth and fi fth lumbar vertebrae were treated with needle puncture of the annulus fi bro-sis. Hematoxylin and eosin staining showed intact nucleus pulposus and annulus fi brosis in untreated L3–L4 discs, but punctured L4–L5 discs showed fi brillation of the nu-cleus pulposus ( Figure 1 C). The nucleus pulposus could be visually distinguished, by MRI signal intensity, from the annulus fi brosis in control and untreated discs but not in punctured discs ( Figure 1 D).

The volumes of L3–L4 and L4–L5 discs were not sig-nifi cantly different in control animals ( P = 0.51; Table 1 ).

In contrast, the volume of untreated L3–L4 discs (1.148 ± 0.054 cm 3 ) was signifi cantly less ( P = 0.02) than the volume of punctured L4–L5 discs (1.283 ± 0.095 cm 3 ).

Figure 1. The morphology of the intervertebral disc is altered in the annular puncture disc degeneration model. Lateral radiographs (A), shown for a representative rabbit, were used to confi rm the anatomic location of the untreated and punctured discs. Histologic slices were taken through the center of the disc in approximately the same loca-tion as the volumetric slice for MRI (B). Hemotoxylin and eosin histol-ogy (C) and standard MR images (D) are shown for the untreated and punctured discs of a representative treatment rabbit.





L3–L4 discs differed from those of punctured L4–L5 discs, with most of the difference in GAG concentration ob-served in the nucleus pulposus ( Figure 5 B).

Displacement and Strain Patterns Altered displacement patterns were observed in punc-tured discs compared with controls ( Figure 2 ). While no signifi cant differences were observed between untreated L3–L4 and punctured L4–L5 groups when displacements were averaged over all specimens in the loading ( P = 0.60) and transverse directions ( P = 0.55), displacement pat-terns varied for each specimen depending on the treatment( Figure 2 B). For most control and untreated L3–L4 speci-mens, but not punctured L4–L5 specimens, displacement patterns showed that, under cyclic compression, the nucle-us pulposus displaced transversely with respect to the sur-rounding annulus fi brosis.

Strains, as computed from smoothed displacement fi elds, were also heterogeneous throughout the discs ( Figure 3 ). However, the averaged strain values within untreated L3–L4 and treated L4–L5 discs were not signifi cantly different (axial, P = 0.41; transverse, P = 0.59; shear, P = 0.40). Mostly ten-sile and some compressive strains were observed in the direc-tion transverse to loading, but predominantly, compressive strains were seen in the loading direction. Strains (and dis-placements), normalized through the depth of untreated and punctured discs, showed large variations ( Figure 4 ).

T 1 Relaxation and GAG Concentration T 1 maps indicated changes in distribution with degen-eration ( Figure 5 A), but T 1 averaged across the full disc showed no signifi cant differences between untreated L3–L4 and punctured L4–L5 groups ( P = 0.20; Table 1 ). GAG concentration in punctured L4–L5 discs was signifi cantly less than that in the untreated L3–L4 discs ( P = 0.015; Table 1 ). In addition, the GAG distributions of untreated

Figure 2. Displacement patterns measured by using DENSE-FSE vary within the same animal and change with degeneration. Displacements within the control discs vary between the L3–L4 and L4–L5 motion segments (A). In the treatment animal, displacement of the nucleus pulposus is distinct from that of the annulus fi brosis in the untreated but not the punctured disc (B).

TABLE 1. Changes in Disc Volume, Spin-Lattice Relaxation Time (T1), and GAG Content After Disc Degeneration

Estimated Whole Disc

Volume (mm 2 )

Spin-Lattice

Relaxation Time (ms)

Glycosami-noglycan

Content ( μ g/mL)

Control L3–L4 (mean ± SEM)

898 ± 83 1043 ± 270 388 ± 111

Control L4–L5 (Mean ± SEM)

988 ± 74 1221 ± 74 252 ± 24

Paired t test(n = 3)

P = 0.51 P = 0.55 P = 0.26

Untreated L3–L4 (Mean ± SEM)

1148 ± 54 1213 ± 108 301 ± 41

Punctured L4–L5 (Mean ± SEM)

1283 ± 95 1397 ± 31 162 ± 10

Paired t test(n = 8)

P = 0.02 P = 0.20 P = 0.015

Whole-disc volume was estimated by using MRI from adjacent image slices through the full intervertebral disc. T 1 was measured by using standard MRI relaxometry techniques, and glycosaminoglycan content was determined by using dGEMRIC. There was a signifi cant difference in volume and glycos-aminoglycan content between the untreated L3–L4 and punctured L4–L5 groups.





and L4/L5 discs, natural differences in size between L3/L4 discs and L4/L5 discs, which are more caudal, could con-tribute to the measured difference in the punctured discs. The puncture degeneration model, especially with larger-diameter needles, 19 results in ruptured fi bers in the annulus, which can affect its ability to resist internal swelling pressure. Therefore, the increase in disc volume could result from the reduced ability of the annulus to resist disc pressure. In addi-tion, although not statistically signifi cant, the slight increase in T 1 may also indicate that more free water is present in the punctured discs ( Table 1 , Figure 5 A). Previous studies show either no change 20 or a decrease 16 in T 1 with advanced disc degeneration. However, degeneration progressed slowly with this model, 6 and animals were killed after only 4 weeks, so tissue swelling might occur early in the degenerative process.

Along with changes in disc morphology and biochemical components, altered mechanical behavior is seen with aging and degenerated discs. 21 Changes to the pressure profi le within the nucleus pulposus have previously been measured, 22 and loading patterns in the periphery of the disc are also altered with degeneration. 23 The puncture model of disc degenera-tion results in decreased compressive stiffness of the disc 19 and

DISCUSSION This study characterized mechanical and biochemical changes in a rabbit annular puncture model. Disc degen-eration was confi rmed by changes in morphology ob-served in histology and with standard MRI. Displacement-encoded MRI was used to demonstrate differences in mechanical behavior between intact and punctured discs. Although average displacement and strain values showed no signifi cant difference between untreated and punctured discs across multiple animals, displacement patterns were altered with annular puncture within each animal. MRI showed a slight decrease in T 1 , and Gd-DTPA 2 − contrast was used to quantify the signifi cant decrease in GAG con-tent with degeneration.

Disc morphology and volume are altered in degenerated discs. Histologic staining showed that fi brillation of the nu-cleus pulposus occurred in punctured discs, which is in agree-ment with previous studies of the rabbit annular puncture model. 5 , 6 In this study, punctured L4/L5 discs had a signifi -cantly higher disc volume than untreated L3/L4 discs after annular puncture ( Table 1 ). Although unexpectedly there were no signifi cant differences between the control L3/L4

Figure 3. Strain patterns in the transverse and loading directions and in shear vary within representative control ( A ) and treatment ( B ) animals. Strains patterns are derived from smoothed displacement fi elds and, as such, appear similar between untreated and punctured discs.





segment, the disc itself experienced bending because of the ge-ometry of discs in relation to the loading axis. Although every effort was made to make the parallel cuts of the vertebral bod-ies orthogonal to the middle plane of the intervertebral disc, an approximation of the angle of the discs showed that the discs in this study were − 0.22 ° ± 5.40 ° off from perpendicular to the loading axis. Because of the range in both the displace-ment and strain values ( Figure 4 ), comparisons between discs

allows for further degeneration of the disc. 19 , 24 In this study, displacements of the nucleus pulposus in most control and un-treated discs, but not treatment discs, were distinct from those of the annulus fi brosis. Previous studies have reported that the nucleus pulposus translates in response to bending loads in the direction opposite to bending. 25 Because transverse displace-ment of the nucleus pulposus was observed, it is likely that, although only compressive loads were applied to the motion

Figure 4. Displacements and strains were normalized through depth and averaged across all discs. Averaged displacements show similar trends in untreated and punc-tured discs. Green-Lagrange strains are com-puted from smoothed displacements and then normalized and averaged. Standard er-ror bars indicate variability in displacement and strain values within the same treatment group.

Figure 5. Spin-lattice relaxation time (T 1 ) and glycosaminoglycan (GAG) distribution, as determined by using dGEMRIC, altered with annular puncture. T 1 maps (A) and GAG content (B) in the unloaded state are shown for a representative control and treatment animal. T 1 decreases and GAG is lost with degeneration in punctured L4–L5 discs.





be considered. Use of gadolinium contrast to quantify GAG content requires that Gd-DTPA 2 − be able to penetrate evenly into the tissues of interest. 36 For administration in vivo , the route of penetration of contrast agent into the disc is through the vertebral body endplate. 37 However, with age and degen-eration, diffusion through the endplate into the interverterbral disc may be affected. 37 , 38 This limits the penetration of gado-linium into the disc 38 , 39 and, therefore, the interpretation of in vivo dGEMRIC data. 36 However, in this study, the discs were submerged in the Gd-DTPA 2 − solution and were allowed ad-equate time (9 hours) for contrast agent penetration before im-aging. Although studies have shown that T 1 after gadolinium administration can correlate with degeneration, 16 the design of future studies must also consider other quantitative MRI techniques that have shown promise for imaging disc degen-eration. Spin–spin relaxation time (T 2 ) can be correlated to water content in the disc, 40 to collagen content, 20 and, in highly degenerated discs, to proteoglycan content. 41 In addition, stud-ies of spin-lattice relaxation time in the rotating frame (T 1 ρ ) in degenerated discs have shown good correlation with proteo-glycan content, 42 GAG content, 10 and also water content and swelling pressure. 43 Nonetheless, the mechanism of correlation between contrast agent concentration and GAG content has been well established for dGEMRIC. 14 , 36

In summary, this study evaluated morphologic, biochemi-cal, and mechanical changes in a rabbit annular puncture disc degeneration model by using histology, standard MRI, dGEMRIC, and DENSE-FSE. Overall GAG content de-creased in treatment discs, with loss of localization of GAG in the nucleus pulposus. Displacement of the nucleus pulposus was distinct in control and untreated discs but not in treat-ment discs. In addition, comparisons of mechanical behavior should be performed on a per-animal basis because variabil-ity in displacements and strains between animals may be on the same order as changes within a single animal. Despite current limitations, a combination of quantitative MRI tech-niques like dGEMRIC and DENSE-FSE can provide valu-able information about changes that occur in early stages and throughout the progression of disc degeneration.

in the same animal ( Figures 2 and 3) may be more useful than generalizations based on a group. Smoothing of displacement data to estimate strain fi elds can also lead to the loss of detail. In addition, the displacement and strain patterns should be ex-amined with consideration of the unique loading environment. It is also pertinent to note that, because discs are affected by an adjacent degenerated disc, 26 the untreated L3–L4 disc may not be representative of completely healthy discs.

In this study, GAG concentration in the punctured discs was signifi cantly decreased from that of untreated discs ( Table 1 ). GAG content increases from the annulus toward the nucleus in normal discs. 27 Herein, GAG content in the nucleus decreased with degeneration, but degenerated discs showed minimal change in GAG in the annulus fi brosis ( Figure 5 ). Proteoglycan loss in early disc degeneration has previously been shown to occur predominantly in the nucleus pulposus rather than the annulus fi brosis. 3 , 27 , 28 Loss of proteoglycans in the nucleus with degeneration 29 , 30 explains the fl attening of the profi le of GAG through the width of the disc that has been reported in previ-ous studies. 27 The loss of GAG content can also be in part due to herniation of the nucleus pulposus during or immediately after surgery.

This study uses quantitative MRI for the evaluation of the biochemical and mechanical outcome of a rabbit punc-ture model of disc degeneration. Because MRI is noninvasive, these methods are translatable to in vivo use for longitudinal studies of disc degeneration models and various interventions. In order for the transition to in vivo use to occur, however, several limitations of these techniques must be considered, as outlined subsequently.

Displacement-encoded MRI allows for the characterization of displacements and strains in the degenerated disc, but the DENSE-FSE technique retains several technical limitations. Be-cause the precision of strains is limited by the signal-to-noise ratio of the phase data, 12 higher signal-to-noise ratio would per-mit less destructive fi ltering of displacement fi elds and prevent loss of detail. In addition, long imaging times make the current DENSE-FSE technique impractical for in vivo use. Neverthe-less, using MRI to determine mechanical behavior of degener-ated discs retains an advantage due to its noninvasive nature. Current methods require invasive procedures to access the disc for measurement of mechanical properties, 31 are feasible only in an ex vivo environment, or disrupt the loading environment of the disc. 25 MRI has previously been used to measure nominal displacements and strains in discs under compression 32 and tor-sion, 33 correlate GAG content to disc mechanics, 34 and deter-mine strains using texture correlation. 35 Displacement-encoded MRI, however, provides displacement data for every pixel of interest, with no bias in computed displacements and strains and no interpolation. 12 Data from such a technique allows for the evaluation of the spatial dependence of mechanical behav-ior in tissues with heterogeneous material properties. Future work in displacement-encoded MRI should aim to produce even faster imaging times, higher signal-to-noise ratio, and im-proved denoising techniques during image processing.

The use of quantitative MRI techniques in correlation with biochemical components also has several limitations that must

➢ Key Points

Biochemical and mechanical changes with annular puncture of rabbit lumbar intervertebral discs were evaluated by using DENSE-FSE and dGEMRIC.

Morphologic changes in the disc degeneration model were confi rmed by using histology and MRI.

Displacements determined by using DENSE-FSE and calculated strains demonstrated that the motion of the nucleus pulposus was distinct from that of the annulus fi brosis for untreated but not for punctured discs.

Imaging with dGEMRIC showed a decreased localiza-tion of GAGs in the nucleus pulposus and an overall decrease in average GAG content in punctured discs compared with untreated discs.





References 1. Luo X , Pietrobon R , Sun SX , et al. Estimates and patterns of direct

health care expenditures among individuals with back pain in the United States . Spine 2004 ; 29 : 79 – 86 .

2. An H , Boden SD , Kang J , et al. Summary statement: Emerging techniques for treatment of degenerative lumbar disc disease . Spine 2003 ; 28 : S24 – 5 .

3. Buckwalter JA . Aging and degeneration of the human interverte-bral disc . Spine 1995 ; 20 : 1307 – 14 .

4. Singh K , Masuda K , An HS . Animal models for human disc degen-eration . Spine J 2005 ; 5 : 267S – 79S .

5. Masuda K , Aota Y , Muehleman C , et al. A novel rabbit model of mild, reproducible disc degeneration by an annulus needle puncture: Correlation between the degree of disc injury and radiological and histological appearances of disc degeneration . Spine 2005 ; 30 : 5 – 14 .

6. Sobajima S , Kompel JF , Kim JS , et al. A slowly progressive and reproducible animal model of intervertebral disc degeneration char-acterized by MRI, X-ray, and histology . Spine 2005 ; 30 : 15 – 24 .

7. Frobin W , Brinckmann P , Biggemann M , et al. Precision measure-ment of disc height, vertebral height and sagittal plane displacement from lateral radiographic views of the lumbar spine . Clin Biomech (Bristol, Avon ) 1997 ; 12 ( suppl 1 ): S1 – S63 .

8. Yoon SH , Miyazaki M , Hong SW , et al. A porcine model of inter-vertebral disc degeneration induced by annular injury characterized with magnetic resonance imaging and histopathological fi ndings. Laboratory investigation . J Neurosurg Spine 2008 ; 8 : 450 – 7 .

9. Luoma K , Vehmas T , Riihimaki H , et al. Disc height and signal intensity of the nucleus pulposus on magnetic resonance imaging as indicators of lumbar disc degeneration . Spine 2001 ; 26 : 680 – 6 .

10. Johannessen W , Auerbach JD , Wheaton AJ , et al. Assessment of human disc degeneration and proteoglycan content using T1rho-weighted magnetic resonance imaging . Spine 2006 ; 31 : 1253 – 7 .

11. Perry J , Haughton V , Anderson PA , et al. The value of T2 relax-ation times to characterize lumbar intervertebral disks: Preliminary results . AJNR Am J Neuroradiol 2006 ; 27 : 337 – 42 .

12. Neu CP , Walton JH . Displacement encoding for the measurement of cartilage deformation . Magn Reson Med 2008 ; 59 : 149 – 55 .

13. Chan DD , Neu CP , Hull ML . Articular cartilage deformation de-termined in an intact tibiofemoral joint by displacement-encoded imaging . Magn Reson Med 2009 ; 61 : 989 – 93 .

14. Bashir A , Gray ML , Boutin RD , et al. Glycosaminoglycan in ar-ticular cartilage: In vivo assessment with delayed Gd(DTPA)(2-)-enhanced MR imaging . Radiology 1997 ; 205 : 551 – 8 .

15. Huang TS , Zucherman JF , Hsu KY , et al. Gadopentetate dimeglu-mine as an intradiscal contrast agent . Spine 2002 ; 27 : 839 – 43 .

16. Niinimäki JL , Parviainen O , Ruohonen J , et al. In vivo quantifi ca-tion of delayed gadolinium enhancement in the nucleus pulposus of human intervertebral disc . J Magn Reson Imaging 2006 ; 24 : 796 – 800 .

17. Noury F , Mispelter J , Szeremeta F , et al. MRI methodological devel-opment of intervertebral disc degeneration: A rabbit in vivo study at 9.4 T . Magn Reson Imaging 2008 ; 26 : 1421 – 32 .

18. Neu CP , Hull ML . Toward an MRI-based method to measure non-uniform cartilage deformation: An MRI-cyclic loading apparatus system and steady-state cyclic displacement of articular cartilage under compressive loading . J Biomech Eng 2003 ; 125 : 180 – 8 .

19. Elliott DM , Yerramalli CS , Beckstein JC , et al. The effect of relative needle diameter in puncture and sham injection animal models of degeneration . Spine 2008 ; 33 : 588 – 96 .

20. Antoniou J , Mwale F , Demers CN , et al. Quantitative magnetic resonance imaging of enzymatically induced degradation of the nucleus pulposus of intervertebral discs . Spine 2006 ; 31 : 1547 – 54 .

21. Iatridis JC , Setton LA , Weidenbaum M , et al. Alterations in the mechanical behavior of the human lumbar nucleus pulposus with degeneration and aging . J Orthop Res 1997 ; 15 : 318 – 22 .

22. Adams MA , McNally DS , Dolan P . “Stress” distributions inside intervertebral discs. The effects of age and degeneration . J Bone Joint Surg Br 1996 ; 78 : 965 – 72 .

23. Frei H , Oxland TR , Rathonyi GC , et al. The effect of nucleoto-my on lumbar spine mechanics in compression and shear loading . Spine 2001 ; 26 : 2080 – 9 .

24. Hsieh AH , Hwang D , Ryan DA , et al. Degenerative annular changes induced by puncture are associated with insuffi ciency of disc biome-chanical function . Spine 2009 ; 34 : 998 – 1005 .

25. Tsantrizos A , Ito K , Aebi M , et al. Internal strains in healthy and degenerated lumbar intervertebral discs . Spine 2005 ; 30 : 2129 – 37 .

26. Kim YE , Goel VK , Weinstein JN , et al. Effect of disc degeneration at one level on the adjacent level in axial mode . Spine 1991 ; 16 : 331 – 5 .

27. Antoniou J , Steffen T , Nelson F , et al. The human lumbar interverte-bral disc: Evidence for changes in the biosynthesis and denaturation of the extracellular matrix with growth, maturation, ageing, and degeneration . J Clin Invest 1996 ; 98 : 996 – 1003 .

28. Pearce RH , Grimmer BJ , Adams ME . Degeneration and the chemi-cal composition of the human lumbar intervertebral disc . J Orthop Res 1987 ; 5 : 198 – 205 .

29. Lipson SJ , Muir H . Experimental intervertebral disc degeneration: Morphologic and proteoglycan changes over time . Arthritis Rheum 1981 ; 24 : 12 – 21 .

30. Zhou H , Hou S , Shang W , et al. A new in vivo animal model to create intervertebral disc degeneration characterized by MRI, ra-diography, CT/discogram, biochemistry, and histology . Spine 2007 ; 32 : 864 – 72 .

31. Southern EP , Fye MA , Panjabi MM , et al. Disc degeneration: A hu-man cadaveric study correlating magnetic resonance imaging and quantitative discomanometry . Spine 2000 ; 25 : 2171 – 5 .

32. Wisleder D , Smith MB , Mosher TJ , et al. Lumbar spine me-chanical response to axial compression load in vivo . Spine 2001 ; 26 : E403 – 9 .

33. Fazey PJ , Song S , Monsas S , et al. An MRI investigation of interver-tebral disc deformation in response to torsion . Clin Biomech (Bristol, Avon) 2006 ; 21 : 538 – 42 .

34. Boxberger JI , Sen S , Yerramalli CS , et al. Nucleus pulposus gly-cosaminoglycan content is correlated with axial mechanics in rat lumbar motion segments . J Orthop Res 2006 ; 24 : 1906 – 15 .

35. O’Connell GD , Johannessen W , Vresilovic EJ , et al. Human inter-nal disc strains in axial compression measured noninvasively using magnetic resonance imaging . Spine 2007 ; 32 : 2860 – 8 .

36. Burstein D , Velyvis J , Scott KT , et al. Protocol issues for delayed Gd(DTPA)(2-)-enhanced MRI (dGEMRIC) for clinical evaluation of articular cartilage . Magn Reson Med 2001 ; 45 : 36 – 41 .

37. Bibby SR , Jones DA , Lee RB , et al. The pathophysiology of the intervertebral disc . Joint Bone Spine 2001 ; 68 : 537 – 42 .

38. Rajasekaran S , Babu JN , Arun R , et al. ISSLS prize winner: a study of diffusion in human lumbar discs: A serial magnetic resonance imaging study documenting the infl uence of the endplate on diffu-sion in normal and degenerate discs . Spine 2004 ; 29 : 2654 – 67 .

39. Nguyen-Minh C , Haughton VM , Papke RA , et al. Measuring diffusion of solutes into intervertebral disks with MR imaging and paramagnetic contrast medium . AJNR Am J Neuroradiol 1998 ; 19 : 1781 – 4 .

40. Liess C , Lusse S , Karger N , et al. Detection of changes in cartilage water content using MRI T2-mapping in vivo . Osteoarthritis Cartilage 2002 ; 10 : 907 – 13 .

41. Benneker LM , Heini PF , Anderson SE , et al. Correlation of ra-diographic and MRI parameters to morphological and biochemi-cal assessment of intervertebral disc degeneration . Eur Spine J 2005 ; 14 : 27 – 35 .

42. Akella SV , Regatte RR , Gougoutas AJ , et al. Proteoglycan-induced changes in T1rho-relaxation of articular cartilage at 4T . Magn Reson Med 2001 ; 46 : 419 – 23 .

43. Nguyen A , Johannessen W , Yoder J , et al. noninvasive quantifi -cation of human nucleus pulposus pressure with use of T1 {rho}-weighted magnetic resonance imaging . J Bone Joint Surg Am 2008 ; 90 : 796 – 802 .


Documents

Mechanical Deformation and Glycosaminoglycan Content ...cneu/NeuLab/Publications_files/ChanDD_etal_2011b.pdfDegeneration Model Eight skeletally mature, male New Zealand white rabbits