- Email: [email protected]
Effects of aging and spinal degeneration on mechanical properties of lumbar supraspinous and interspinous ligaments
The Spine Journal 2 (2002) 95–100
Clinical study
Effects of aging and spinal degeneration on mechanical properties of lumbar supraspinous and interspinous ligaments Takahiro Iida, MD, Kuniyoshi Abumi, MD*, Yoshihisa Kotani, MD, Kiyoshi Kaneda, MD Department of Orthopaedic Surgery, Hokkaido University Graduate School of Medicine, Kita-15, Nishi-7, Kitaku, Sapporo, 060-8638, Japan Received 14 June 2001; accepted 10 December 2001
Abstract
Background context: The effects of aging and spinal degeneration on the mechanical properties of spinal ligaments are still unknown, although there have been several studies demonstrating those of normal spinal ligaments. Purpose: To investigate the mechanical properties of the human posterior spinal ligaments in human lumbar spine, and their relation to age and spinal degeneration parameters. Study design/setting: Destructive uniaxial tensile tests were performed on the human supraspinous and interspinous ligaments at L4–5 level. Their mechanical properties were compared with age and spinal degeneration using several imaging modalities. Patient sample: Twenty-four patients with lumbar degenerative diseases on whom posterior surgeries were performed, with the age ranging from 18 to 85 years. Outcome measures: The ultimate load and elastic stiffness as structural properties, the degree of disc degeneration, range of segmental motion, the disc height, disc space narrowing ratio and degree of facet degeneration as the parameters of spinal degeneration. Methods: Twenty-four supraspinous and interspinous ligaments at the L4–5 level were obtained from posterior surgeries of patients with lumbar degenerative disease. The mechanical tests of boneligament-bone complexes were performed in a uniaxial tensile fashion with a specially designed clamp device. The ultimate load and elastic stiffness were calculated as structural properties. The degree of disc degeneration, range of segmental motion, the disc height, disc space narrowing ratio and degree of facet degeneration were examined by using radiographs, computed tomography and magnetic resonance imaging. Results: The average and SD value of ultimate load, elastic stiffness, tensile strength and elastic modulus were 203102.9 N, 60.636.7 N/mm, 1.20.6 Mpa and 3.32.1 Mpa, respectively. A significant negative correlation was found between age and tensile strength (p 0.02). The specimens with facet degeneration showed lower values in tensile strength and elastic modulus than those without facet degeneration (p0.04). However, no correlation was found between disc-related parameters and tensile strength. Conclusions: The mechanical strength of human lumbar posterior spinal ligaments decreases with age and facet degeneration, particularly in the ligament substance. © 2002 Elsevier Science Inc. All rights reserved.
Keywords:
Biomechanics; Lumbar spine; Posterior spinal ligament; Age factors
Introduction FDA device/drug status: not applicable. Nothing of value received from a commercial entity related to this research. * Corresponding author. Kuniyoshi Abumi, MD, Department of Orthopaedic Surgery, Hokkaido University School of Medicine, Kita-15, Nishi-7, Kitaku, Sapporo, 060-8638 Hokkaido, Japan. Tel.: 81-11-706-6054 (ext. 5934); fax: 81-11-706-6054. E-mail address: [email protected] (K. Abumi)
Although there have been several studies demonstrating the mechanical properties of normal spinal ligaments, the effects of aging and spinal degeneration on the mechanical properties of spinal ligaments are still unknown. The remodeling occurs in the normal ligaments according to the mechanical demand of the region, as in the bone. Maximum stress and stiffness of the ligaments decreases when the
1529-9430/02/$ – see front matter © 2002 Elsevier Science Inc. All rights reserved. PII: S1529-9430(02)00 1 4 2 - 0
96
T. Iida et al. / The Spine Journal 2 (2002) 95–100
region is fixed [1,2] and increases when hard exercise is carried out [3]. Kotani et al. [1] demonstrated decreased mechanical properties of the posterior spinal ligaments after posterior spinal instrumentation and fusion in an animal model. Aging of the ligaments has the same effects as the fixation of the region. Noyes and Grood [4] determined the mechanical properties of anterior cruciate bone-ligament-bone (BLB) specimens from humans. They found significant reductions in strength and stiffness of ligament units to occur with advancing age to a greater degree than expected. With regard to the spinal ligaments, biochemical, morphological and mechanical changes according to age were reported in the anterior and posterior longitudinal ligament [5–7] and in the ligamentum flavum [8,9]. Nachemson and Evans [9] established stress–strain curves for the ligamentum flavum subjected to traction at the L3–4 level, determining the values of stress, strain and modulus of elasticity. They found decreasing values with age [9]. With regard to the supraspinous ligament (SSL) and interspinous ligament (ISL), several studies about mechanical properties were reported. However, no study was carried out to determine the relation between mechanical properties of SSL and ISL, and their aging or degeneration [10–15]. The degeneration of the spinal motion unit is considered to have three phases: initial dysfunction phase, instability phase and restabilization phase [16]. The intervertebral disc and zygoapophyseal joint (facet joint) degenerate and interact with each other to create each phase [16]. The environment of altered stress under the degeneration phases would have influenced the mechanical properties of the spinal ligaments. Also, the degeneration of the ligament itself would have influenced them. Because the most posterior surgeries of the lumbar spine are performed to the degenerative diseases, understanding of the mechanical properties of the degenerated or aged posterior ligaments is essential in order to decide whether to spare the ligaments. The purpose of this study was to investigate changes in the mechanical properties of the SSL and ISL of the human lumbar spine in relation to age and spinal degeneration determined by several imaging modalities. Materials and methods
Table 1 Diagnosis and age at posterior lumbar surgeries Diseases
Number
Age (Average years)
Lumbar canal stenosis Degenerative spondylolisthesis Lumbar disc herniation Degenerative lumbar scoliosis Total
14 7 2 1 24
62–79 (69.9) 52–85 (67.3) 18–41 (29.5) 80 18–85 (66.2)
means of Ringers solution and high-humidity environments. They were then carefully dissected to leave only the SSL, ISL and adjacent spinous processes. These two ligaments were tested in a combined manner because of the impossibility of distinguishing them [1,12,17]. The width of SSL was decided as the same width as adjacent spinous processes, because it was not possible to distinguish the SSL from the lumbosacral fascia. The depth of the SSL and ISL combined was set to 20 mm, as measured from the dorsal surface of the SSL at three different levels (Fig. 1). Measurement of cross-sectional area The cross-sectional areas of SSL and ISL combined were measured using the special apparatus with charge coupled device camera and video dimension analyzer [18] presented in Fig. 2. The width of the center of the ligament was measured in 36 directions with rotating the stepping motor 5 degrees each. The data were put into DeltaGraf Pro 3 (DeltaPoint, Inc., Bellevue, Washington, USA) to show the cross-sectional view. Then the cross-sectional view was put into NIH Image (National Institutes of Health, Bethesda, Maryland) to calculate the cross-sectional area. Although we have not performed a calibration procedure, the cross-sectional area of a specimen was 225.75 3.06 mm2 and the coefficient of variation was 1.4%. This result might be sufficient for this study. Measurement of initial length Initial length of the SSL and ISL combined was measured as the shortest distance between both spinous processes on the X-ray film taken before mechanical testing,
Specimen tested Twenty four SSLs and ISLs at L4–5 level were obtained from posterior surgeries of patients with lumbar degenerative diseases ranging from 18 to 85 years of age (Table 1). The BLB complexes were obtained with about the halves of both adjacent spinous processes. The dissected BLBs were double-wrapped in plastic bags and kept frozen at 30°C. During these initial preparations, some muscle tissues were left on the BLB complex to protect the ligaments from contact with the air. Preparation of BLB specimens Before testing, the BLBs were thawed at room temperature. The preparation was kept moist throughout the test by
Fig. 1. The schematic of bone-ligament-bone specimen. ISL interspinous ligament; SP the half of the adjacent spinous process; SSL supraspinous ligament.
T. Iida et al. / The Spine Journal 2 (2002) 95–100
97
and the lower end plate of L5. The disc height and disc space narrowing ratio were measured by the particular method showed in Fig. 3. The disc space narrowing ratio means the ratio of the disc height of the referred disc (L4–5) to the maximum value among the disc height of lower three discs (L3–4, L4–5, L5–S). The degree of facet degeneration was classified into four groups from computed tomography and oblique view of radiographs showed in Table 2. Statistical analysis The correlation analysis was performed between these parameters using Spearman’s correlation analysis and Fisher’s protected least significant difference. Fig. 2. Special apparatus for measurement of cross-sectional area. The width of the center of the ligament was measured by charge coupled device (CCD) camera and video dimension analyzer in 36 directions with stepping motor. BLB bone-ligament-bone complex.
Results
applying a tensile load by connecting a weight (300 g). The measurement accuracy of the initial length was 0.1 mm.
The average and SD values of ultimate load and elastic stiffness were 203.2102.9 N and 60.636.7 N/mm, respectively. The cross-sectional area and initial length of the SSL and ISL combined were 185.158.6 mm2 and 9.94.0 mm. The tensile strength and elastic modulus were 1.20.6 MPa and 3.32.1 Mpa, respectively.
Destructive tensile testing The mechanical tests of BLB were performed in a uniaxial tensile fashion using the MTS 858 Bionix Test System (MTS System Company, Minneapolis, MN) with a specially designed clamp device controlled by a personal computer (PC 9801 RA, NEC Company, Tokyo, Japan). A tensile pre-load of 3 N was applied followed by testing at a constant displacement rate of 0.5 mm per second until failure. The load and displacement data were acquired at 10 Hz to produce a load-displacement curve. Mechanical parameters The ultimate load (N) and elastic stiffness (N/mm) were calculated as structural properties based on the load-displacement curves obtained. Elastic stiffness (N/mm) was defined by the slope of elastic zone in the load-displacement curve. Stress-strain curves were obtained based on the calculated stress, which was the tensile load divided by the cross-sectional area, and the calculated strain, which was the displacement divided by the initial length. The tensile strength (MPa) and elastic modulus (MPa) were determined from the stress-strain curves. Elastic modulus (MPa) was defined by the slope of the linear portion in the stress-strain curve.
Mechanical tests
Spinal degeneration The degree of disc degeneration in Kellgren’s classification and DeCandido’s classification were shown in Table 3 with the degree of facet degeneration in our classification. Grade 2 was most frequent for the disc degeneration in Kellgren’s classification, Grade 3 in DeCandido’s classification and Grade 4 for the facet degeneration in our classification. The range of segmental motion was 6.93.7 degrees, and the disc space narrowing ratio was 2626%. Correlation analysis The significant negative correlation was found between age and tensile strength (p0.02, R20.21, Fig. 4) and be-
Imaging modalities Several parameters of spinal degeneration were examined from radiographs, computed tomography and magnetic resonance imaging. The degree of disc degeneration was classified into four groups based on Kellgren’s classification [19] on radiographs and DeCandido’s classification [20] on magnetic resonance imaging, respectively. The range of segmental motion was calculated from functional lateral radiographs, defined as the range of angulation between the upper end plate of L4
Fig. 3. The method for measuring disc height and disc space narrowing ratio. X is the cross point of two lines along the inferior and superior vertebral border (XAXB, XPXQ). The disc height H(ABPQ)/2AP. The disc space narrowing ratio (%)(HmHo)/Hm100. Hm is the maximum value among the disc height of lower three discs (L3–4, L4–5, L5–S). Ho is the disc height of the referred disc (L4–5).
98
T. Iida et al. / The Spine Journal 2 (2002) 95–100
Table 2 The degree of facet degeneration Findings on CT and radiograph Grade
Narrowing
Irregularity
Sclerosis
1 2 3 4
() in either finding () in either finding
CT computed tomography.
tween age and elastic modulus (p0.02, R2 0.23, Fig. 4). The specimens with facet degeneration beyond Grade 2 showed lower values in tensile strength and elastic modulus than those without facet degeneration (p0.04, Fig. 5, A). The same trend was found in elastic stiffness and ultimate load without any significance (Fig. 5, B and C). However, no correlation was found between disc-related parameters and tensile strength (R20.001, Fig. 6). Also no correlation was found among range of segmental motion, disc height, disc height narrowing ratio and tensile strength and among those and elastic modulus. Although disc degeneration classified on Kellgren’s and DeCandido’s classification showed no correlation with age, facet degeneration showed weak correlation with it. Positive correlation was found between disc degeneration and facet degeneration (R20.20, Fig. 7). The ultimate load and elastic stiffness showed no correlation with several imaging parameters. Displacement at failure revealed decreasing value according to age.
Fig. 4. Correlation between age and tensile strength and between age and elastic modulus. The negative correlation was found between both of them (p.02).
that with this method of acquiring specimens from posterior surgeries of patients with lumbar degenerative disease, most of the data were gathered from older subjects. Although the majority of the patients with lumbar disc herniations were 30 to 50 years of age, SSL and ISL were rarely sacrificed, because fusion operations are rarely performed. The authors recognize that the conclusions are still preliminary, and further replication of the findings in a larger and more scat-
Discussion This study was designed to clarify whether age or any imaging parameters of segmental motion and degeneration are related to mechanical properties of posterior spinal ligaments. Mechanical properties of the SSL and ISL vary according to their level [14]. Breaking load of the SSL was maximum at the L2–3 or L3–4 level, and lesser at the L4–5 level [14]. To exclude these variations, 24 L4–5 specimens alone were studied and analyzed in this study. Besides the acquisition of SSL and ISL from posterior surgeries of patients with lumbar degenerative diseases enable us to collect various imaging modalities in vivo of the specimens. Potential limitations of this study may be, first, the small number of specimens. The majority of the data was clustered around 60 to 80 years of age, and there were very few samples under 50 years of age. It was inevitable, however,
Table 3 The degree of disc and facet degeneration Grade
Kellgren’s classification
DeCandido’s classification
Our classification of facet degeneration
1 2 3 4
4 11 7 2
0 6 14 4
2 5 5 12
Fig. 5. Correlation between facet degeneration and mechanical parameters: (A) tensile strength and elastic modulus, (B) elastic stiffness, (C) ultimate load. The specimens with facet degeneration beyond Grade 2 showed values in tensile strength and elastic modulus lower than those without facet degeneration (p.04). No significant correlation was found in elastic stiffness and ultimate load.
T. Iida et al. / The Spine Journal 2 (2002) 95–100
Fig. 6. Correlation between disc degeneration and tensile strength. No correlation was found between them.
tered sample would be worthwhile. Second, the direction of the applied load during the testing was not the same as the fiber direction for the ISL but was same as that for the SSL. It is difficult to apply the load along the same direction as the ISL. The BLB complex was aligned to make its position the same as the neutral position in vitro. Because mechanical strength of the SSL has been shown to be much stronger than that of the ISL by Myklebust et al. [14] and Rissanen [17], data obtained in this testing represented a true loaddeformation data of the ligaments. The results presented in this article were similar to the results of other authors, although other studies differed in the methods of the measurement and the level and age of the specimens. Myklebust et al. [14] tested the SSL and ISL separately at the L4–5 level and demonstrated failure loads of 329161 N and 11576 N, respectively. Chazal et al. [12] observed nearly the same values (80 to 300 N), testing the SSL together with the ISL. Adams et al. [10] measured the original length of 13.46.3 mm and estimated the crosssectional area of 17153 mm2 for the SSL and ISL in the lumbar spine. These comparisons demonstrate the validity of the method in this study.
Fig. 7. Correlation between disc degeneration and facet degeneration. Positive correlation was found between them.
99
A significant negative correlation was found between age and tensile strength and between age and elastic modulus without any correlation between age and ultimate load. This suggests that aging influences the quality of the ligament, leading to a decrease in ligament strength. It appears to be because of hypertrophy of the ligaments that the ultimate load does not decrease parallel to age with some materials. The significant difference in tensile strength and elastic modulus between the specimens with facet degeneration beyond Grade 2 and those without facet degeneration raises two possibilities: the segmental immobilization resulted from degeneration, or aging itself decreased the ligament strength. However, most segments preserved motion, and no correlation was found between range of segmental motion and mechanical parameters of ligaments. This suggests that the former factor cannot explain our results. On the other hand, facet degeneration showed weak correlation with age. This suggests that the latter factor can explain our results. These findings cannot be extrapolated to the general population, because this study included only problematic patients. Considering the properties of posterior ligaments only in the patients with lumbar degenerative diseases, however, these findings would be worthwhile. The authors recognize that these correlations do not show causality, just association. Age may be a surrogate for some other process. This may be why the correlations between various parameters and age showed in the results are very weak, and why the correlations between elastic stiffness and facet degeneration and between ultimate load and facet degeneration showed no significance, showing the same trend as elastic modulus and tensile strength. Further studies are necessary to investigate whether this process occurs in patients with lumbar degenerative disease. The intervertebral discs and facet joints may interactively degenerate, causing the degeneration of the spinal unit as a whole. The discs of the patients with lumbar canal stenosis and degenerative spondylolisthesis, whose facets degenerate severely, also show degeneration to some degree. The facets of the patients with lumbar disc herniations, whose discs degenerate severely, however, do not always degenerate. Although the ratio of the diseases included in the study has some effect on the results, this may be one of the reasons why disc-related parameters showed no correlation with age and mechanical properties of the posterior ligaments. Postacchini et al. [8] pointed out in their histological study that the ligamenta flava of the patients with disc herniation had morphologic features similar to those of the controls of similar ages. The disc degeneration itself may have less influence on the degeneration of the posterior ligaments. Panjabi et al. [15] obtained three-dimensional flexibility data of the functional spinal unit to demonstrate wider neutral zones for the degenerated spine, indicating a higher probability of overstretching the ligaments in the degenerated spine. The displacement at failure, on the other hand, was shown to
100
T. Iida et al. / The Spine Journal 2 (2002) 95–100
decrease with age in their study. As the initial length of the ligaments was measured attaching with weight of 300 g, the ligaments should be slack in the neutral position with no tension to the ligaments. The degenerated posterior ligaments may increase their length under small tensile force within the physiologic motion, while beyond the physiologic range of motion the increase of length may be small with increasing tension to them. Decreasing involvement of posterior ligaments in spinal stability may accelerate the degeneration of these ligaments, requiring the discs and facets to play an increasing role. The posterior ligaments of the lumbar spine and facets and discs degenerate and interact with each other to create the degeneration of the spinal unit as a whole. At the decompression surgery for lumbar canal stenosis, it is controversial for the interspinous and supraspinous ligaments to be spared or not to be spared in the biomechanical viewpoint. Because this disease is based on the degeneration of the intervertebral discs and zygoapophyseal joints, the interspinous and supraspinous ligaments are thought to be the sites of the degenerative change and decrease in tensile strength. Abumi et al. [21] showed that the division of the SSL and ISL does not affect any lumbar spinal motions in their study of the graded facetectomies in fresh human cadaver with an average age of 46.5 years. Their study suggests that the posterior lumbar ligaments do not have important role in segmental stability of the lumbar spine. Dumas [13] determined the most restrictive ligament to be the ligamentum flavum followed by the articular ligament, ISL and SSL. When the facet joint or articular ligament is preserved in decompression surgery for lumbar degenerative diseases, decreased tensile strength and increased stiffness in the ISLs and SSLs are not critical. These mechanical changes could be the result of histological changes of the ligaments. Rissanen [17] reported in detail on the interspinous and supraspinous ligaments of the lumbar spine with special reference to age-induced changes and ruptures of these ligaments histologically. Metaplasia into fibrocartilage was demonstrated in the supraspinous ligaments after the twentieth year of age and was pronounced by middle age. In the interspinous ligaments, breaking of the fiber bundle, fragmentation, disappearance of stainability, destruction of tendinous tissue and finally, cavity formation occur gradually after the twentieth year of age. Postacchini et al. [8] studied ligamentum flava obtained from lumbar disc herniation and lumbar stenosis at histological, histochemical and ultrastructural levels. Fibrotic changes, chondroid metaplasia and calcification reduce the elasticity of the ligaments. Further histological study should be performed to clarify the relation between degeneration and mechanical properties of the supraspinous and interspinous ligaments of the lumbar spine.
References [1] Kotani Y, Cunningham BW, Cappuccino A, Kaneda K, McAfee PC. The effects of spinal fixation and destabilization on the biomechanical and histologic properties of spinal ligaments. An in vivo study. Spine 1998;23:672–82. [2] Noyes FR. Functional properties of knee ligaments and alterations induced by immobilization: a correlative biomechanical and histological study in primates. Clin Orthop 1977;123:210–42. [3] Tipton C. Influence of exercise on strength of medial collateral ligament of dogs. Am J Physiol 1970;218:894–902. [4] Noyes FR, Grood ES. The strength of the anterior cruciate ligament in human and Rhesus monkeys. Age-related and species-related changes. J Bone Joint Surg (Am) 1976;58:1074–82. [5] Arutynow AJ. Basic problems of the pathology and surgical treatment of prolapsed intervertebral discs. Vopr Neirokhir 1962;4:21–4. [6] Harris RI, MacNab I. Structural changes in the lumbar intervertebral discs. Their relationship to low back pain and sciatica. J Bone Joint Surg (Br) 1954;36:304–22. [7] Tkaczuk H. Tensile properties of human lumbar longitudinal ligaments. Acta Orthop Scand 1968(suppl);115:3–69. [8] Postacchini F, Gumina S, Cinotti G, Perugia D, DeMartino C. Ligamenta flava in lumbar disc herniation and spinal stenosis. Light and electron microscopic morphology. Spine 1994;19:917–22. [9] Nachemson AL, Evans JH. Some mechanical properties of the third human lumbar interlaminar ligament (ligamentum flavum). J Biomech 1968;1:211–20. [10] Adams MA, Hutton WC, Stott JRR. The resistance of flexion of the lumbar intervertebral joint. Spine 1980;5:245–53. [11] Asano S, Kaneda K, Umehara S, Tadano S. The mechanical properties of the human L4–5 functional spinal units during cyclic loading. Spine 1992;17:1343–52. [12] Chazal J, Tanguy A, Bourges M, et al. Biomechanical properties of spinal ligaments and a histological study of the supraspinal ligament in traction. J Biomech 1985;18:167–76. [13] Dumas GA, Beaudoin L, Drouin G. In situ mechanical behavior of posterior spinal ligaments in the lumbar region. An in vitro study. J Biomech 1987;20:301–10. [14] Myklebust JB, Pintar F, Yoganandan N, et al. Tensile strength of spinal ligaments. Spine 1988;13:526–31. [15] Panjabi MM, Goel VK, Takata K. Physiologic strains in the lumbar spinal ligaments. An in situ biomechanical study. Spine 1982;7:192–203. [16] Kirkaldy-Willis WH, Wedge JH, Yong-Hing K, Reilly J. Pathology and pathogenesis of lumbar spondylosis and stenosis. Spine 1978;3:319–28. [17] Rissanen PM. The surgical anatomy and pathology of the supraspinous and interspinous ligaments of the lumbar spine with special reference to ligament ruptures. Acta Orthop Scand 1960(suppl);46:5–100. [18] Yamamoto E, Hayashi K, Yamamoto N. Mechanical properties of collagen fascicles from the rabbit patellar tendon. J Biomech Eng 1999;121:124–31. [19] Kellgren JH. Atlas of standard radiographs of arthritis. In: The epidemiology of chronic rheumatism. Oxford: Black Well Scientific Publication, 1963:63–74. [20] DeCandido P, Reinig JW, Dwyer AJ, Thompson KJ, Ducker TB. Magnetic resonance assessment of the distribution of lumbar spine disc degeneration changes. J Spinal Disord 1988;1:9–15. [21] Abumi K, Panjabi MM, Kramen KM, Duranceau J, Oxland T, Crisco JJ. Biomechanical evaluation of lumbar spinal stability after graded facetectomies. Spine 1990;15:1142–7.
Clinical study
Effects of aging and spinal degeneration on mechanical properties of lumbar supraspinous and interspinous ligaments Takahiro Iida, MD, Kuniyoshi Abumi, MD*, Yoshihisa Kotani, MD, Kiyoshi Kaneda, MD Department of Orthopaedic Surgery, Hokkaido University Graduate School of Medicine, Kita-15, Nishi-7, Kitaku, Sapporo, 060-8638, Japan Received 14 June 2001; accepted 10 December 2001
Abstract
Background context: The effects of aging and spinal degeneration on the mechanical properties of spinal ligaments are still unknown, although there have been several studies demonstrating those of normal spinal ligaments. Purpose: To investigate the mechanical properties of the human posterior spinal ligaments in human lumbar spine, and their relation to age and spinal degeneration parameters. Study design/setting: Destructive uniaxial tensile tests were performed on the human supraspinous and interspinous ligaments at L4–5 level. Their mechanical properties were compared with age and spinal degeneration using several imaging modalities. Patient sample: Twenty-four patients with lumbar degenerative diseases on whom posterior surgeries were performed, with the age ranging from 18 to 85 years. Outcome measures: The ultimate load and elastic stiffness as structural properties, the degree of disc degeneration, range of segmental motion, the disc height, disc space narrowing ratio and degree of facet degeneration as the parameters of spinal degeneration. Methods: Twenty-four supraspinous and interspinous ligaments at the L4–5 level were obtained from posterior surgeries of patients with lumbar degenerative disease. The mechanical tests of boneligament-bone complexes were performed in a uniaxial tensile fashion with a specially designed clamp device. The ultimate load and elastic stiffness were calculated as structural properties. The degree of disc degeneration, range of segmental motion, the disc height, disc space narrowing ratio and degree of facet degeneration were examined by using radiographs, computed tomography and magnetic resonance imaging. Results: The average and SD value of ultimate load, elastic stiffness, tensile strength and elastic modulus were 203102.9 N, 60.636.7 N/mm, 1.20.6 Mpa and 3.32.1 Mpa, respectively. A significant negative correlation was found between age and tensile strength (p 0.02). The specimens with facet degeneration showed lower values in tensile strength and elastic modulus than those without facet degeneration (p0.04). However, no correlation was found between disc-related parameters and tensile strength. Conclusions: The mechanical strength of human lumbar posterior spinal ligaments decreases with age and facet degeneration, particularly in the ligament substance. © 2002 Elsevier Science Inc. All rights reserved.
Keywords:
Biomechanics; Lumbar spine; Posterior spinal ligament; Age factors
Introduction FDA device/drug status: not applicable. Nothing of value received from a commercial entity related to this research. * Corresponding author. Kuniyoshi Abumi, MD, Department of Orthopaedic Surgery, Hokkaido University School of Medicine, Kita-15, Nishi-7, Kitaku, Sapporo, 060-8638 Hokkaido, Japan. Tel.: 81-11-706-6054 (ext. 5934); fax: 81-11-706-6054. E-mail address: [email protected] (K. Abumi)
Although there have been several studies demonstrating the mechanical properties of normal spinal ligaments, the effects of aging and spinal degeneration on the mechanical properties of spinal ligaments are still unknown. The remodeling occurs in the normal ligaments according to the mechanical demand of the region, as in the bone. Maximum stress and stiffness of the ligaments decreases when the
1529-9430/02/$ – see front matter © 2002 Elsevier Science Inc. All rights reserved. PII: S1529-9430(02)00 1 4 2 - 0
96
T. Iida et al. / The Spine Journal 2 (2002) 95–100
region is fixed [1,2] and increases when hard exercise is carried out [3]. Kotani et al. [1] demonstrated decreased mechanical properties of the posterior spinal ligaments after posterior spinal instrumentation and fusion in an animal model. Aging of the ligaments has the same effects as the fixation of the region. Noyes and Grood [4] determined the mechanical properties of anterior cruciate bone-ligament-bone (BLB) specimens from humans. They found significant reductions in strength and stiffness of ligament units to occur with advancing age to a greater degree than expected. With regard to the spinal ligaments, biochemical, morphological and mechanical changes according to age were reported in the anterior and posterior longitudinal ligament [5–7] and in the ligamentum flavum [8,9]. Nachemson and Evans [9] established stress–strain curves for the ligamentum flavum subjected to traction at the L3–4 level, determining the values of stress, strain and modulus of elasticity. They found decreasing values with age [9]. With regard to the supraspinous ligament (SSL) and interspinous ligament (ISL), several studies about mechanical properties were reported. However, no study was carried out to determine the relation between mechanical properties of SSL and ISL, and their aging or degeneration [10–15]. The degeneration of the spinal motion unit is considered to have three phases: initial dysfunction phase, instability phase and restabilization phase [16]. The intervertebral disc and zygoapophyseal joint (facet joint) degenerate and interact with each other to create each phase [16]. The environment of altered stress under the degeneration phases would have influenced the mechanical properties of the spinal ligaments. Also, the degeneration of the ligament itself would have influenced them. Because the most posterior surgeries of the lumbar spine are performed to the degenerative diseases, understanding of the mechanical properties of the degenerated or aged posterior ligaments is essential in order to decide whether to spare the ligaments. The purpose of this study was to investigate changes in the mechanical properties of the SSL and ISL of the human lumbar spine in relation to age and spinal degeneration determined by several imaging modalities. Materials and methods
Table 1 Diagnosis and age at posterior lumbar surgeries Diseases
Number
Age (Average years)
Lumbar canal stenosis Degenerative spondylolisthesis Lumbar disc herniation Degenerative lumbar scoliosis Total
14 7 2 1 24
62–79 (69.9) 52–85 (67.3) 18–41 (29.5) 80 18–85 (66.2)
means of Ringers solution and high-humidity environments. They were then carefully dissected to leave only the SSL, ISL and adjacent spinous processes. These two ligaments were tested in a combined manner because of the impossibility of distinguishing them [1,12,17]. The width of SSL was decided as the same width as adjacent spinous processes, because it was not possible to distinguish the SSL from the lumbosacral fascia. The depth of the SSL and ISL combined was set to 20 mm, as measured from the dorsal surface of the SSL at three different levels (Fig. 1). Measurement of cross-sectional area The cross-sectional areas of SSL and ISL combined were measured using the special apparatus with charge coupled device camera and video dimension analyzer [18] presented in Fig. 2. The width of the center of the ligament was measured in 36 directions with rotating the stepping motor 5 degrees each. The data were put into DeltaGraf Pro 3 (DeltaPoint, Inc., Bellevue, Washington, USA) to show the cross-sectional view. Then the cross-sectional view was put into NIH Image (National Institutes of Health, Bethesda, Maryland) to calculate the cross-sectional area. Although we have not performed a calibration procedure, the cross-sectional area of a specimen was 225.75 3.06 mm2 and the coefficient of variation was 1.4%. This result might be sufficient for this study. Measurement of initial length Initial length of the SSL and ISL combined was measured as the shortest distance between both spinous processes on the X-ray film taken before mechanical testing,
Specimen tested Twenty four SSLs and ISLs at L4–5 level were obtained from posterior surgeries of patients with lumbar degenerative diseases ranging from 18 to 85 years of age (Table 1). The BLB complexes were obtained with about the halves of both adjacent spinous processes. The dissected BLBs were double-wrapped in plastic bags and kept frozen at 30°C. During these initial preparations, some muscle tissues were left on the BLB complex to protect the ligaments from contact with the air. Preparation of BLB specimens Before testing, the BLBs were thawed at room temperature. The preparation was kept moist throughout the test by
Fig. 1. The schematic of bone-ligament-bone specimen. ISL interspinous ligament; SP the half of the adjacent spinous process; SSL supraspinous ligament.
T. Iida et al. / The Spine Journal 2 (2002) 95–100
97
and the lower end plate of L5. The disc height and disc space narrowing ratio were measured by the particular method showed in Fig. 3. The disc space narrowing ratio means the ratio of the disc height of the referred disc (L4–5) to the maximum value among the disc height of lower three discs (L3–4, L4–5, L5–S). The degree of facet degeneration was classified into four groups from computed tomography and oblique view of radiographs showed in Table 2. Statistical analysis The correlation analysis was performed between these parameters using Spearman’s correlation analysis and Fisher’s protected least significant difference. Fig. 2. Special apparatus for measurement of cross-sectional area. The width of the center of the ligament was measured by charge coupled device (CCD) camera and video dimension analyzer in 36 directions with stepping motor. BLB bone-ligament-bone complex.
Results
applying a tensile load by connecting a weight (300 g). The measurement accuracy of the initial length was 0.1 mm.
The average and SD values of ultimate load and elastic stiffness were 203.2102.9 N and 60.636.7 N/mm, respectively. The cross-sectional area and initial length of the SSL and ISL combined were 185.158.6 mm2 and 9.94.0 mm. The tensile strength and elastic modulus were 1.20.6 MPa and 3.32.1 Mpa, respectively.
Destructive tensile testing The mechanical tests of BLB were performed in a uniaxial tensile fashion using the MTS 858 Bionix Test System (MTS System Company, Minneapolis, MN) with a specially designed clamp device controlled by a personal computer (PC 9801 RA, NEC Company, Tokyo, Japan). A tensile pre-load of 3 N was applied followed by testing at a constant displacement rate of 0.5 mm per second until failure. The load and displacement data were acquired at 10 Hz to produce a load-displacement curve. Mechanical parameters The ultimate load (N) and elastic stiffness (N/mm) were calculated as structural properties based on the load-displacement curves obtained. Elastic stiffness (N/mm) was defined by the slope of elastic zone in the load-displacement curve. Stress-strain curves were obtained based on the calculated stress, which was the tensile load divided by the cross-sectional area, and the calculated strain, which was the displacement divided by the initial length. The tensile strength (MPa) and elastic modulus (MPa) were determined from the stress-strain curves. Elastic modulus (MPa) was defined by the slope of the linear portion in the stress-strain curve.
Mechanical tests
Spinal degeneration The degree of disc degeneration in Kellgren’s classification and DeCandido’s classification were shown in Table 3 with the degree of facet degeneration in our classification. Grade 2 was most frequent for the disc degeneration in Kellgren’s classification, Grade 3 in DeCandido’s classification and Grade 4 for the facet degeneration in our classification. The range of segmental motion was 6.93.7 degrees, and the disc space narrowing ratio was 2626%. Correlation analysis The significant negative correlation was found between age and tensile strength (p0.02, R20.21, Fig. 4) and be-
Imaging modalities Several parameters of spinal degeneration were examined from radiographs, computed tomography and magnetic resonance imaging. The degree of disc degeneration was classified into four groups based on Kellgren’s classification [19] on radiographs and DeCandido’s classification [20] on magnetic resonance imaging, respectively. The range of segmental motion was calculated from functional lateral radiographs, defined as the range of angulation between the upper end plate of L4
Fig. 3. The method for measuring disc height and disc space narrowing ratio. X is the cross point of two lines along the inferior and superior vertebral border (XAXB, XPXQ). The disc height H(ABPQ)/2AP. The disc space narrowing ratio (%)(HmHo)/Hm100. Hm is the maximum value among the disc height of lower three discs (L3–4, L4–5, L5–S). Ho is the disc height of the referred disc (L4–5).
98
T. Iida et al. / The Spine Journal 2 (2002) 95–100
Table 2 The degree of facet degeneration Findings on CT and radiograph Grade
Narrowing
Irregularity
Sclerosis
1 2 3 4
() in either finding () in either finding
CT computed tomography.
tween age and elastic modulus (p0.02, R2 0.23, Fig. 4). The specimens with facet degeneration beyond Grade 2 showed lower values in tensile strength and elastic modulus than those without facet degeneration (p0.04, Fig. 5, A). The same trend was found in elastic stiffness and ultimate load without any significance (Fig. 5, B and C). However, no correlation was found between disc-related parameters and tensile strength (R20.001, Fig. 6). Also no correlation was found among range of segmental motion, disc height, disc height narrowing ratio and tensile strength and among those and elastic modulus. Although disc degeneration classified on Kellgren’s and DeCandido’s classification showed no correlation with age, facet degeneration showed weak correlation with it. Positive correlation was found between disc degeneration and facet degeneration (R20.20, Fig. 7). The ultimate load and elastic stiffness showed no correlation with several imaging parameters. Displacement at failure revealed decreasing value according to age.
Fig. 4. Correlation between age and tensile strength and between age and elastic modulus. The negative correlation was found between both of them (p.02).
that with this method of acquiring specimens from posterior surgeries of patients with lumbar degenerative disease, most of the data were gathered from older subjects. Although the majority of the patients with lumbar disc herniations were 30 to 50 years of age, SSL and ISL were rarely sacrificed, because fusion operations are rarely performed. The authors recognize that the conclusions are still preliminary, and further replication of the findings in a larger and more scat-
Discussion This study was designed to clarify whether age or any imaging parameters of segmental motion and degeneration are related to mechanical properties of posterior spinal ligaments. Mechanical properties of the SSL and ISL vary according to their level [14]. Breaking load of the SSL was maximum at the L2–3 or L3–4 level, and lesser at the L4–5 level [14]. To exclude these variations, 24 L4–5 specimens alone were studied and analyzed in this study. Besides the acquisition of SSL and ISL from posterior surgeries of patients with lumbar degenerative diseases enable us to collect various imaging modalities in vivo of the specimens. Potential limitations of this study may be, first, the small number of specimens. The majority of the data was clustered around 60 to 80 years of age, and there were very few samples under 50 years of age. It was inevitable, however,
Table 3 The degree of disc and facet degeneration Grade
Kellgren’s classification
DeCandido’s classification
Our classification of facet degeneration
1 2 3 4
4 11 7 2
0 6 14 4
2 5 5 12
Fig. 5. Correlation between facet degeneration and mechanical parameters: (A) tensile strength and elastic modulus, (B) elastic stiffness, (C) ultimate load. The specimens with facet degeneration beyond Grade 2 showed values in tensile strength and elastic modulus lower than those without facet degeneration (p.04). No significant correlation was found in elastic stiffness and ultimate load.
T. Iida et al. / The Spine Journal 2 (2002) 95–100
Fig. 6. Correlation between disc degeneration and tensile strength. No correlation was found between them.
tered sample would be worthwhile. Second, the direction of the applied load during the testing was not the same as the fiber direction for the ISL but was same as that for the SSL. It is difficult to apply the load along the same direction as the ISL. The BLB complex was aligned to make its position the same as the neutral position in vitro. Because mechanical strength of the SSL has been shown to be much stronger than that of the ISL by Myklebust et al. [14] and Rissanen [17], data obtained in this testing represented a true loaddeformation data of the ligaments. The results presented in this article were similar to the results of other authors, although other studies differed in the methods of the measurement and the level and age of the specimens. Myklebust et al. [14] tested the SSL and ISL separately at the L4–5 level and demonstrated failure loads of 329161 N and 11576 N, respectively. Chazal et al. [12] observed nearly the same values (80 to 300 N), testing the SSL together with the ISL. Adams et al. [10] measured the original length of 13.46.3 mm and estimated the crosssectional area of 17153 mm2 for the SSL and ISL in the lumbar spine. These comparisons demonstrate the validity of the method in this study.
Fig. 7. Correlation between disc degeneration and facet degeneration. Positive correlation was found between them.
99
A significant negative correlation was found between age and tensile strength and between age and elastic modulus without any correlation between age and ultimate load. This suggests that aging influences the quality of the ligament, leading to a decrease in ligament strength. It appears to be because of hypertrophy of the ligaments that the ultimate load does not decrease parallel to age with some materials. The significant difference in tensile strength and elastic modulus between the specimens with facet degeneration beyond Grade 2 and those without facet degeneration raises two possibilities: the segmental immobilization resulted from degeneration, or aging itself decreased the ligament strength. However, most segments preserved motion, and no correlation was found between range of segmental motion and mechanical parameters of ligaments. This suggests that the former factor cannot explain our results. On the other hand, facet degeneration showed weak correlation with age. This suggests that the latter factor can explain our results. These findings cannot be extrapolated to the general population, because this study included only problematic patients. Considering the properties of posterior ligaments only in the patients with lumbar degenerative diseases, however, these findings would be worthwhile. The authors recognize that these correlations do not show causality, just association. Age may be a surrogate for some other process. This may be why the correlations between various parameters and age showed in the results are very weak, and why the correlations between elastic stiffness and facet degeneration and between ultimate load and facet degeneration showed no significance, showing the same trend as elastic modulus and tensile strength. Further studies are necessary to investigate whether this process occurs in patients with lumbar degenerative disease. The intervertebral discs and facet joints may interactively degenerate, causing the degeneration of the spinal unit as a whole. The discs of the patients with lumbar canal stenosis and degenerative spondylolisthesis, whose facets degenerate severely, also show degeneration to some degree. The facets of the patients with lumbar disc herniations, whose discs degenerate severely, however, do not always degenerate. Although the ratio of the diseases included in the study has some effect on the results, this may be one of the reasons why disc-related parameters showed no correlation with age and mechanical properties of the posterior ligaments. Postacchini et al. [8] pointed out in their histological study that the ligamenta flava of the patients with disc herniation had morphologic features similar to those of the controls of similar ages. The disc degeneration itself may have less influence on the degeneration of the posterior ligaments. Panjabi et al. [15] obtained three-dimensional flexibility data of the functional spinal unit to demonstrate wider neutral zones for the degenerated spine, indicating a higher probability of overstretching the ligaments in the degenerated spine. The displacement at failure, on the other hand, was shown to
100
T. Iida et al. / The Spine Journal 2 (2002) 95–100
decrease with age in their study. As the initial length of the ligaments was measured attaching with weight of 300 g, the ligaments should be slack in the neutral position with no tension to the ligaments. The degenerated posterior ligaments may increase their length under small tensile force within the physiologic motion, while beyond the physiologic range of motion the increase of length may be small with increasing tension to them. Decreasing involvement of posterior ligaments in spinal stability may accelerate the degeneration of these ligaments, requiring the discs and facets to play an increasing role. The posterior ligaments of the lumbar spine and facets and discs degenerate and interact with each other to create the degeneration of the spinal unit as a whole. At the decompression surgery for lumbar canal stenosis, it is controversial for the interspinous and supraspinous ligaments to be spared or not to be spared in the biomechanical viewpoint. Because this disease is based on the degeneration of the intervertebral discs and zygoapophyseal joints, the interspinous and supraspinous ligaments are thought to be the sites of the degenerative change and decrease in tensile strength. Abumi et al. [21] showed that the division of the SSL and ISL does not affect any lumbar spinal motions in their study of the graded facetectomies in fresh human cadaver with an average age of 46.5 years. Their study suggests that the posterior lumbar ligaments do not have important role in segmental stability of the lumbar spine. Dumas [13] determined the most restrictive ligament to be the ligamentum flavum followed by the articular ligament, ISL and SSL. When the facet joint or articular ligament is preserved in decompression surgery for lumbar degenerative diseases, decreased tensile strength and increased stiffness in the ISLs and SSLs are not critical. These mechanical changes could be the result of histological changes of the ligaments. Rissanen [17] reported in detail on the interspinous and supraspinous ligaments of the lumbar spine with special reference to age-induced changes and ruptures of these ligaments histologically. Metaplasia into fibrocartilage was demonstrated in the supraspinous ligaments after the twentieth year of age and was pronounced by middle age. In the interspinous ligaments, breaking of the fiber bundle, fragmentation, disappearance of stainability, destruction of tendinous tissue and finally, cavity formation occur gradually after the twentieth year of age. Postacchini et al. [8] studied ligamentum flava obtained from lumbar disc herniation and lumbar stenosis at histological, histochemical and ultrastructural levels. Fibrotic changes, chondroid metaplasia and calcification reduce the elasticity of the ligaments. Further histological study should be performed to clarify the relation between degeneration and mechanical properties of the supraspinous and interspinous ligaments of the lumbar spine.
References [1] Kotani Y, Cunningham BW, Cappuccino A, Kaneda K, McAfee PC. The effects of spinal fixation and destabilization on the biomechanical and histologic properties of spinal ligaments. An in vivo study. Spine 1998;23:672–82. [2] Noyes FR. Functional properties of knee ligaments and alterations induced by immobilization: a correlative biomechanical and histological study in primates. Clin Orthop 1977;123:210–42. [3] Tipton C. Influence of exercise on strength of medial collateral ligament of dogs. Am J Physiol 1970;218:894–902. [4] Noyes FR, Grood ES. The strength of the anterior cruciate ligament in human and Rhesus monkeys. Age-related and species-related changes. J Bone Joint Surg (Am) 1976;58:1074–82. [5] Arutynow AJ. Basic problems of the pathology and surgical treatment of prolapsed intervertebral discs. Vopr Neirokhir 1962;4:21–4. [6] Harris RI, MacNab I. Structural changes in the lumbar intervertebral discs. Their relationship to low back pain and sciatica. J Bone Joint Surg (Br) 1954;36:304–22. [7] Tkaczuk H. Tensile properties of human lumbar longitudinal ligaments. Acta Orthop Scand 1968(suppl);115:3–69. [8] Postacchini F, Gumina S, Cinotti G, Perugia D, DeMartino C. Ligamenta flava in lumbar disc herniation and spinal stenosis. Light and electron microscopic morphology. Spine 1994;19:917–22. [9] Nachemson AL, Evans JH. Some mechanical properties of the third human lumbar interlaminar ligament (ligamentum flavum). J Biomech 1968;1:211–20. [10] Adams MA, Hutton WC, Stott JRR. The resistance of flexion of the lumbar intervertebral joint. Spine 1980;5:245–53. [11] Asano S, Kaneda K, Umehara S, Tadano S. The mechanical properties of the human L4–5 functional spinal units during cyclic loading. Spine 1992;17:1343–52. [12] Chazal J, Tanguy A, Bourges M, et al. Biomechanical properties of spinal ligaments and a histological study of the supraspinal ligament in traction. J Biomech 1985;18:167–76. [13] Dumas GA, Beaudoin L, Drouin G. In situ mechanical behavior of posterior spinal ligaments in the lumbar region. An in vitro study. J Biomech 1987;20:301–10. [14] Myklebust JB, Pintar F, Yoganandan N, et al. Tensile strength of spinal ligaments. Spine 1988;13:526–31. [15] Panjabi MM, Goel VK, Takata K. Physiologic strains in the lumbar spinal ligaments. An in situ biomechanical study. Spine 1982;7:192–203. [16] Kirkaldy-Willis WH, Wedge JH, Yong-Hing K, Reilly J. Pathology and pathogenesis of lumbar spondylosis and stenosis. Spine 1978;3:319–28. [17] Rissanen PM. The surgical anatomy and pathology of the supraspinous and interspinous ligaments of the lumbar spine with special reference to ligament ruptures. Acta Orthop Scand 1960(suppl);46:5–100. [18] Yamamoto E, Hayashi K, Yamamoto N. Mechanical properties of collagen fascicles from the rabbit patellar tendon. J Biomech Eng 1999;121:124–31. [19] Kellgren JH. Atlas of standard radiographs of arthritis. In: The epidemiology of chronic rheumatism. Oxford: Black Well Scientific Publication, 1963:63–74. [20] DeCandido P, Reinig JW, Dwyer AJ, Thompson KJ, Ducker TB. Magnetic resonance assessment of the distribution of lumbar spine disc degeneration changes. J Spinal Disord 1988;1:9–15. [21] Abumi K, Panjabi MM, Kramen KM, Duranceau J, Oxland T, Crisco JJ. Biomechanical evaluation of lumbar spinal stability after graded facetectomies. Spine 1990;15:1142–7.