The immediate effect of repeated loading on the ...

Knee Surg Sports Traumatol Arthrosc (2010) 18:694–701 DOI 10.1007/s00167-009-1001-z

SPORTS MEDICINE

The immediate effect of repeated loading on the compressive strength of young porcine lumbar spine Olof Thoreson • Adad Baranto • Lars Ekstro¨m Sten Holm • Mikael Hellstro¨m • Leif Swa¨rd

•

Received: 21 September 2009 / Accepted: 13 November 2009 / Published online: 9 December 2009 Ó Springer-Verlag 2009

Abstract The human spine is exposed to repeated loading during daily activities and more extremely during sports. Despite this, there remains a lack of knowledge regarding the immediate effects on the spine due to this mode of loading. Age-specific spinal injury patterns has been demonstrated and this implies differences in reaction to load mode and load history The purpose of the present study was to investigate the impact of cyclic pre-loading on the biomechanical properties and fracture patterns of the adolescent spine in an experimental model. Eight functional spinal units from four young porcine spines were harvested. The functional spinal units were cyclic loaded with 20,000 cycles and then axially compressed to failure. The compression load at failure, ultimate stress and viscoelastic parameters were calculated. The functional spinal units were examined with plain radiography, computer tomography and MRI before and after the loading, and finally macroscopically and histologically. The median compression load at failure in this study was 8.3 kN (range 5.6–8.7 kN). The median deformation for all cases was 2.24 mm (range 2.30–2.7 mm) and stiffness was 3.45 N/mm (range 3.5–4.5 N/mm). A fracture was seen on

O. Thoreson A. Baranto (&) L. Ekstro¨m S. Holm L. Swa¨rd Department of Orthopaedics, The Sahlgrenska Academy at Gothenburg University and Sahlgrenska University Hospital, 413 45 Gothenburg, Sweden e-mail: [email protected] M. Hellstro¨m Department of Radiology, The Sahlgrenska Academy at Gothenburg University and Sahlgrenska University Hospital, 413 45 Gothenburg, Sweden

123

radiograph in one case, on CT and macroscopically in seven, and on MRI and histologically in all eight cases. The cyclic loaded functional spinal units in the present study were not more sensitive to axial compression than noncyclic loaded functional spinal units from young porcine. The endplate and the growth zone were the weakest part in the cyclic loaded functional spinal units. Disc signal reduction and disc height reduction was found on MRI. The E-modulus value found in this study was of the same order of magnitude as found by others using a porcine animal model. Keywords Porcine Cyclic loading Compression Intervertebral disc Athletes

Introduction The human spine is exposed to repeated loading during normal daily activity and particularly in sports of varying intensities, e.g. long distance running or cross-country motor-bike racing. Unfortunately, there is still a lack of knowledge in the literature about the effects of various types of submaximal loading such as cyclic loading on the spine. Increased spinal loading during adolescent, originated from sports activity has been demonstrated to increase the risk of spinal injury and indicate that the growing spine is susceptible to trauma [3, 4, 17, 20, 26, 27]. There are some experimental studies focusing on the failure mechanisms of the adolescent lumbar spine when subjected to single loading-to-failure conditions, showing that the growing immature spine exhibits specific failure properties which differ from those of the adult spine [1, 2, 18]. However, there are no studies that focus on the immediate effect of vibration induced

Knee Surg Sports Traumatol Arthrosc (2010) 18:694–701

pre-loading on spinal strength of the adolescent, neither human nor experimental. The spine of athletes is exposed for different cyclic loading and quantities, e.g. a soccer game, cross-country motor-bike racing game, tennis game, gymnastics game or a long distance running race. The soccer players, for example, run different stretches depending on their position in the football field; a midfielder moves about 20 km, of which 7.5 km constitute walking during a 90-min game. During this time, the player does 90–110 jerks with maximal or near maximal loading. A defender moves about 6–7 km during a 90-min game. This means that a midfielder is exposed to 20,500 cyclic loads. An elite marathon runner takes 20,000–25,000 steps during a race, depending on running technique, body length and other individual factors. In humans, the diurnal cycle of erect and supine posture results in large variations in load, i.e. between moving around during the daily activities and lying down during the night. During daily activity, the intervertebral disc is subjected to a combination of static and dynamic loadings [5, 21, 29]. The resulting intervertebral disc fluid exchange is well known from experimental studies. The intervertebral disc stiffness has been shown to increase and the disc height to decrease during static creep loading and superimposed dynamic loading enhance this phenomenon. Similar findings also apply to moderate physiological cyclic loading of the disc [12, 16] and are believed to be the results of fluid extrusion from the nucleus and creep deformation of the annulus fibrosus to a great extent [6, 7]. The mechanical function of the intervertebral disc depends on both elastic and viscoelastic properties, and both must be taken into consideration in experimental studies [15]. Epidemiological studies have shown that the most frequent cause of spinal injury is repeated loading [13]. Exposure to long-term whole-body occupational vibrations has been found to cause harm to the lower lumbar spine [10, 23]. Long-term occupational whole-body vibration may trigger bone remodelling processes in accordance with the Wolff law [25] and leads to various spinal disorders which have been shown in experimental and clinical studies [25]. There are, to our knowledge, no experimental studies published that has studied the strength properties of young lumbar spines as a function of load history. In the present experimental study, it was hypothesized that cyclic loading of functional spinal units (FSUs) prior to axial compression may lead to a decreased spinal strength and thus likely to an increased risk of spinal injury when compared to axial compression without prior cyclic loading. More relevant question is if the spine of athletes is more sensitive for injuries after training or a game?

695

Methods Experimental animals and procedures Four young, healthy, male domestic pigs with an age of 6 months and weight between 65 and 70 kg were used in the present experimental study. In terms of skeletal maturity, these animals correspond to adolescents. The animals were sedated by an intramuscular injection of Ketamine (KetalarÒ 20, 15–20 mg/kg body weight), and then anaesthetized by intravenous injections of Methomidate chloride (HypnodilÒ 21, 3–5 mg/kg body weight) and StresnilÒ 22 (azaperon 0.1 mg/kg body weight). The pigs were killed and the lumbar spines were harvested. The muscles were removed from the lumbar spines, while the posterior bony elements, capsular structures and ligaments were left intact [1, 2]. Eight FSUs, four at the L2–L3 level and four at the L4–L5 level (Table 1), were collected. The FSUs were placed in plastic bags to minimize dehydration and were stored at about ?8°C in a refrigerator between preparation and testing. In order to facilitate mounting of the FSUs, the intervertebral discs from the adjacent levels above and below were removed, i.e. from the superior endplate of the cranial vertebra and from the inferior endplate of the caudal vertebra. The segment height, and the width, anterior-posterior (AP) diameter and height of the disc were measured with a digital calliper. The superior part of the cranial vertebra and the inferior part of the caudal vertebra were then mounted in special testing cups and stabilized with polyester putty. The vertebrae were mounted in such a way so as to achieve parallel surfaces that were perpendicular to the loading axis. This study was approved by the Regional Ethical Review Board. The experiments comply with the current laws in the country. Plain radiography, CT and MRI examinations The FSUs were examined with plain radiography, CT and MRI 1–2 days before and 1–6 h after cyclic loading and axial compression to failure. A 1.0 Tesla Signa Advantage system (GE Medical System) was used for the MRI examinations, using an extremity coil. Sagittal T1 images (spin echo; repetition time [TR], 500; echo time [TE], 26) Table 1 Showing the viscoelastic results Viscoelastic parameters E1 elastic damping modulus

E2 elastic damping constant

g Viscous damping constant

12.3 MPa

3.7 MPa

3.6 GPa

123

696

and T2 images (fast spin echo; TR 3860; TE 101.3), with a field of view of 24 9 12 cm as well as axial T1 (spin echo; TR, 440; TE, 30) and T2 images (fast spin echo; TR, 3000; TE, 104.1) with a field of view of 16 9 8 cm were obtained. A matrix of 512 9 384 and a slice thickness of 3 mm with a 0.5-mm gap were used in all sequences. CT was performed with a multidetector CT machine (GE Lightspeed/GE Healthcare) using a slice thickness of 0.625 mm, speed of 9.37 mm/rotation and a pitch of 0.938:1 at 140 kV and 570 mA. Transverse and sagittal CT reconstructions were used for analysis. Biomechanical tests Specimens were wrapped in saline-soaked gauze to prevent dehydration of the discs. The FSUs were loaded in axial sinusoidal cyclic compression with a force of 0–1,000 N at a frequency of 3 Hz. Each FSU was loaded with a total of 20,000 cycles, which took approximately 2 hours. Immediately after the cyclic loading, the FSUs were exposed to continuous axial compression to failure using a ramp command. The axial compression loading to failure utilized a deformation rate of 5 mm/min. Failure was defined as the event when a more than 5% drop from the peak force value was detected, often accompanied by a audible crack. Ultimate strength was calculated by dividing the force at failure by the intervertebral disc area, which was estimated using an elliptic model, based on the anterior-posterior diameter and width of the disc. The disc height was used as the original value in the calculation of strain, and the E-modulus was derived from data originating from the most linear part of the stress–strain curve. A three-parameter standard linear solid model was utilized in order to give a more detailed description of the vibrocreep phenomena. Approximating a linear system, the sinusoidal pulsatile compression force can be divided into two components, a constant force component and a component with zero mean. This allows for the use of a step load analysis. The mean deformation data originated from the cyclic loading part of the protocol were used in the description of the vibrocreep behaviour of the FSU and further calculations. Under a step load, the strain history of the threeparameter standard linear solid model can be written as eðtÞ 1 1 t ¼ þ 1 e s2 r0 E1 E2 where e (t) is the strain as a function of time t (s), r0 (Pa) is the applied constant stress, E1 (Pa) is the elastic damping modulus and E2 (Pa) is the elastic damping constant, s2 (s) is the relaxation time constant. An additional parameter, the viscous damping constant g (Pa s), can be calculated

123


from g = E2s2. An in-depth description of this model and procedure can be found in [9]. No further statistical evaluation was performed with this data. Macroscopic and histological examinations After the radiological examinations, the FSUs were frozen and stored in a -20°C freezer for at least 12 h. The frozen specimens were then sawed into 3–4 mm thick sagittal slices, using a bench saw. Each slice was macroscopically examined for injuries and digital photographs were taken. The slices were then decalcified, dehydrated, fixed in paraffin and cut into 4-lm thick sections using a microtome. The samples were stained with haematoxylin-eosin and Alcian blue solution [1, 2]. Eight histological sections from each FSU were examined microscopically (magnification 946–250) for injuries, which were later related to the macroscopic and radiological findings. Photographs were taken to document the histological pathology. Definition of injuries The injuries were defined according to protocol by Baranto et al. [1, 2]. Fracture of the endplate was defined as a fracture line through the endplate itself. Separation of the endplate was defined as a widening (fracture) of the growth zone with separation of the endplate from the vertebral body. Statistical analysis The results were compared to a control group, derived from a study of porcine spinal strength by Lundin et al. [19]. Identical test conditions and age- or sex-matched animals as the present study were used, except there was no cyclic loading imposed on the specimens before strength testing. The Mann–Whitney U-test was used for comparison of the biomechanical parameters between the test group and the control group. All differences were tested using the 5% (P \ 0.05) level of statistical significance.

Results There were no anatomical anomalies or traumatic injuries detected on plain radiography, CT or MRI in any case before cyclic loading. Biomechanical values A time-dependent behaviour of the intervertebral disc was noted in all cases; the cyclic loading induced a creep phenomenon, i.e. an increasing deformation with time.


697

Figure 1 presents the mean deformation value for all cases (n = 8) with respect to time. A three-parameter solid model was used to mathematically describe the creep curve. The results from the evaluation are presented in Table 1. Ultimate compression load at failure, deformation and stiffness for all cases are presented in Table 2. Median ultimate load was 8.3 kN (range 5.6–8.7 kN). Specimens 3 and 4, from the same subject, showed a lower compression load at failure than the other specimens. There was no difference between the specimens concerning age, weight and examined FSU levels. There were no statistically significant differences between the specimens exposed to cyclic loading and the control group (median ultimate load 7.9 kN) [18] with respect to ultimate axial compression strength. The mean ultimate compression force at failure for the L2–L3 units was 8.5 kN and for the control group, 8.3 kN and at the L4–L5 it was 7.8 kN and for the control group, 7.5 kN. Median E-modulus was 77 MPa (range 57– 100 MPa). Plain radiographic, CT and MRI examinations A fracture through the endplate (caudal/cranial), growth zone and the dorsal part of the vertebral body was faintly visible in 1 of 8 cases (Table 3) after cyclic loading and axial compression. On CT examination, a fracture was clearly seen, extending through the endplate (caudal/cranial), growth zone and the dorsal part of the vertebral body in 7 of 8 cases (Table 3) after cyclic loading and axial compression (Fig. 2). On MRI, a fracture was found through the endplate (caudal/cranial), growth zone and the dorsal part of the vertebral body in all 8 cases (Table 3). There was slight reduction in disc height and slight to moderate reduction of

Macroscopic and histological examinations In 7 of 8 cases, a fracture could be found at the dorsal margin of the endplate (caudal/cranial), through the epiphyseal plate, through the growth zone and finally through the dorsal corner of the vertebral body (Fig. 4, Table 3). In two cases, a fracture or separation of the endplate from the vertebral body was seen in the caudal vertebra. On microscopic examination of the histological samples, a fracture could be found in all 8 cases at the dorsal margin of the endplate (caudal/cranial), through the epiphyseal plate, through the growth zone and finally through the dorsal corner of the vertebral body (Fig. 5, Table 3). Separations of the growth zones from the vertebral bodies alone were found in 5 of 8 cases. In 5 cases, a fracture of the dorsal margin of the vertebral body or a separation of the growth zone was found anteriorly and in 8 cases posteriorly. In two cases, the fractures of the dorsal margin of the vertebral body or separations of the growth zone were in the cranial vertebra and in 8 cases in the dorsal part of the caudal vertebra.

Discussion The cyclic loaded FSUs in the present study were not more sensitive to axial compression than non-cyclic loaded FSUs in the control group. The endplate and the growth zone were the weakest parts and an injury pattern similar to that of the control group was noted. The vibrocreep characteristics showed a great concurrence between specimens, as can be seen from the small

3.0

2.5

Deformation [mm]

Fig. 1 Illustration of the vibrocreep phenomenon, i.e. Deformation of the disc with time, induced by cyclic loading at a frequency of 3 Hz with 1,000 N peak load. Thick line represents the mean value for all eight specimens and the thin lines represent one standard deviation

the disc signal intensity on T2-weighted images, in 7 of 8 cases (Fig. 3).

2.0

1.5

1.0

0.5

0.0 0

20

40

60

80

100

120

Time [minutes]

123

698


Table 2 Functional spinal units (FSUs) levels, ultimate force at failure, deformation and stiffness for all specimens after cyclic loading, immediately followed by continuous axial compression to failure Subject

Specimen

FSU level

Deformation (mm)

Stiffness (N/mm)

Ultimate force at failure (kN)

1

1

L2–L3

1.83

4.5

8.3

1

2

L4–L5

2.33

3.7

8.7

2 2

3 4

L4–L5 L2–L3

1.94 2.00

2.9 2.8

5.6 5.7

3

5

L2–L3

2.31

3.7

8.6

3

6

L4–L5

2.7

3.0

8.3

4

7

L2–L3

2.5

3.4

8.5

4

8

L4–L5

2.3

3.6

8.4

Table 3 Plain radiography, Computed tomography (CT), Magnetic Resonance Imaging (MRI), macroscopic and histological findings for all cases after cyclic loading, immediately followed by continuous axial compression to failure Specimen

1

Radiological examinations

Macroscopic examination

Histological examination

Plain radiography injury ±

CT injury ±

MRI injury ±

Macroscopic injury ±

Microscopic injury ±

Injury ant/post ±

Injury cranial/caudal vertebra ±

-

?

?

?

?

;

;

2

-

?

?

?

?

?/?

;

3

-

?

?

-

?

;

;

4

-

?

?

?

?

;

;

5

-

?

?

?

?

?/?

?/?

6

-

-

?

?

?

?/?

;

7

-

?

?

?

?

?/?

?/?

8

?

?

?

?

?

?/?

;

Fig. 2 CT after cyclic loading followed by axial compression to failure (Case 8). a In the sagittal plane, a fracture through the endplate, growth zone and the dorsal part of the vertebral body (arrow) is shown. b In the axial plane, the extension of the fracture is seen (arrow)

standard deviation in Fig. 1. Since the animals were biologically very similar, i.e. of the same age, weight, breeding and sex, this finding is not surprising. In previous studies, it has been shown that the creep characteristics are dependent on the structural conditions of the tested specimen [8, 11, 15]. Increasing load levels enhances creep. Therefore, selection of another load level would probably yield a

123

different outcome [15, 28]. Since a thorough analysis of the creep characteristics was not the main goal of the study, the evaluation in the present study was merely descriptive. Cyclic loading (dynamic) has been shown to produce different creep behaviour when compared with constant (static) loading [21]. However, these differences are not fully understood and need to be further investigated.


699

Fig. 5 Histological picture of the same case (Case 8) showing a fracture through the endplate, growth zone and the dorsal part of the vertebral body (white arrow). There is also a fracture or separation of the endplate from the vertebral body in the growth zone (black arrow)

Fig. 3 MRI before and after failure loading (Case 8). a Before cyclic loading. b After cyclic loading followed by axial compression to failure. A fracture through the endplate, growth zone and the dorsal part of the vertebral body is noted (arrow). There is damage to the nucleus pulposus with leakage of nucleus material (white) in the fracture

Fig. 4 Macroscopic photograph of the same case (Case 8) showing a fracture through the endplate, growth zone and the dorsal part of the vertebral body (arrow)

The E-modulus value found in this study was of the same order of magnitude as reported in other studies using porcine animal models [1, 2]. Compared with the noncyclic loaded FSUs, the stress–strain curve lacked, to a great extent, the initial non-linear phase that is generally very noticeable when tests of collagenous structures, such as the disc, are performed. Also, the transition from an elastic region to a plastic region seemed to differ between the two test situations. A notable bend in the stress–strain

can be seen in Fig. 6, and the plastic region appears to be lacking the characteristic non-linear appearance until it reaches a yield point and fails macroscopically. These characteristics were found in a majority of the tested specimens. In the present study, the median compression load at failure was 8.3 kN. The median disc deformation was 2.24 mm (range 2.30–2.7 mm), and the stiffness was 3.45 N/mm (range 3.5–4.5 N/mm). This is in accordance with the axial compression loads at failure for the control group of non-cyclic loaded normal adolescent pigs FSUs (7.9 kN) reported by Lundin et al. [19], where the same experimental set-up and equipment as in this study was used. This means that, within the guidelines of the present study in terms of number of load cycles and load level, the hypothesis that vibrated FSUs are more sensitive for ultimate compression loads was false, since no statistical difference between the vibrated and non-vibrated FSUs was found. However, if the test conditions were different, using higher cyclic load frequency and/or higher load levels, the fatigue limit of the FSUs would probably be reached, and thus the original hypothesis might hold true. Compared to sized- and age-matched degenerated adolescent pig’s spine segments, a significant difference in strength was noted [1, 2]. Almost twice as much force was found to be needed in the degenerated group of specimens compared to the specimens in the present study, under equivalent test conditions, albeit non-cyclic loaded. The loading quantity of 20,000 cycles was chosen in order to roughly correspond to the number of repeated loads a human spine will be subjected to during for example, a soccer game or a long distance running race. The soccer players run different stretches depending on their position in the football field; a midfielder moves

123

700


Fig. 6 Axial compressive strength test of specimen 5 after cyclic loading (thick line) when compared to a typical case from the reference material without prior cyclic loading (dashed line; Lundin et al. [19])

about 20 km, of which 7.5 km constitute walking during a 90-minute game. During this time, the player does 90–110 jerks with maximal or near maximal loading. A defender moves about 6–7 km during a 90-min game. One step is about 0.5 m. This means that a midfielder walks 2 9 7,500 m, making a total of 15 km and runs for 5.5 km. This means that a midfielder is exposed to 20,500 repetitive loads. An elite marathon runner takes 20,000–25,000 steps during a race, depending on running technique, body length and other individual factors. It can be speculated that he outcome in a comparison between controls and cyclic loaded FSUs strength may be different if another loading mode is use, such as flexion and extension loading. The present study is, in conception, similar to some previously published studies, i.e. it is a single experimental condition protocol for a specific quantity of cyclic loads [11, 15, 24, 28], although the number of cycles in the present study was considerably higher. The transfer to a clinical situation of results from these experimental studies may be difficult since human daily activity, especially in sports, includes a complex pattern of motion and a long-term loading exposure. The result of the simplified test protocol might underestimate the risk of reaching the fatigue limits in real life activity and, as a consequence, underestimate the risk for injury to the spine. The FSUs were investigated with plain radiography, CT and MRI, and examined macroscopically and histologically in search for injuries. A fracture was seen on plain radiography in one case, on CT and macroscopically in seven, and on MRI and histogically in all eight cases. Fractures were located at the dorsal margin of the endplate, through the epiphyseal plate, through the growth zone and finally through the dorsal corner of the vertebral body. The fractures in the present study showed the same patterns as control group [19] and in degenerated FSUs from young

123

porcine spines [2]. Thus, the endplates and the growth zones were the weakest parts in the axially compressed FSUs in the present study. The apparent weakness of the vertebral endplates has previously been reported in experimental studies on adult FSUs [1, 2, 14, 18, 22]. The present study also used CT for examination of the injuries. To our knowledge, CT has not been used previously in experimental studies on young pigs or human spines. CT is an excellent tool for diagnosing fractures in the spine, but due to the radiation it is not ethically acceptable to use it on young healthy volunteers in human studies. CT proved to be almost equal to MRI in diagnosing fractures in the endplate and vertebral body. However, fractures or separations of the endplate in the growth zone could not be seen on CT, while major injuries could easily be found on MRI. Therefore, MRI is best suited for diagnosing injuries in the growth zone and it is also the method of choice for studies in humans [1, 2]. There are several limitations in this study; for example it is an experimental study and only the FSU are used. The muscles and soft tissue also have an important role in the mechanical behaviour of the spine. The load is only performed in axial compression and probably additional information may be getting from bending loads. Another limitation is that the performed pre-cyclic load for 20,000 cycles is a short episode of loading and also on single loading condition, therefore, more amount of cycles and higher load than 3 Hz may geld in different results. The FSUs were not sensitive for axial compression in cyclic pre-loaded spines from young pigs. But we found that there were disc signal and disc height reduction and on MRI after loading. Despite that the FSUs were not sensitive it might be another result if the FSUs were pre-cycled with more amounts of cycles or higher loads. This would in that case gives as other information and that there is an increased risk for injuries in sports. Top athletes are training several


times and days per week with an additional game. Therefore, there is no time for recovery. Therefore, the spine of young athletes is exposed for considerably high loads with high risks for injuries and overuse injuries that are causing degenerations findings. To prevent spinal injuries, the most important factor is to avoid high loads on the spine of athletes before the spine is fully grown, which means at least after the age of 18. It is also important to avoid extreme bending movements of the spine in sports when it is possible. The amount of training is of fundamental importance to reduce to avoid overuse injuries on the spine in athletes. The age of attending competition should also be high then it is now. The young athletes ought to concentrate on coordination, techniques and balance before the growth spurt.

Conclusions The cyclic loaded FSUs in the present study were not more sensitive to axial compression than non-cyclic loaded FSUs from young pigs. The endplate and the growth zone were the weakest parts in the cyclic loaded FSUs. Disc signal reduction and disc height reduction were found on MRI. The E-modulus value found in this study was of the same order of magnitude as found by others using a porcine animal model. Acknowledgments The authors acknowledge the financial support of The Medical Society of Gothenburg, Sweden, The Research Council of the Swedish Sports Confederation, Swedish Society of Spinal Surgeons/4S, Anna and Edwin Bergs Foundation and Government grants under the LUA/ALF agreement. The authors thank radiology technologist Pa¨r-Arne Svensson, Department of Radiology at Queen Silvia’s Children’s Hospital, Sahlgrenska University Hospital, Gothenburg, Sweden.

References 1. Baranto A, Ekstro¨m L, Hellstro¨m M et al (2005) Fracture patterns of the adolescent porcine spine: an experimental loading study in bending-compression. Spine 30:75–82 2. Baranto A, Ekstro¨m L, Holm S et al (2005) Vertebral fractures and separations of endplates after traumatic loading of adolescent porcine spines with experimentally-induced disc degeneration. Clin Biomech (Bristol, Avon) 20:1046–1054 3. Baranto A, Hellstro¨m M, Cederlund CG et al (2009) Back pain and MRI changes in the thoraco-lumbar spine of top athletes in four different sports: a 15-year follow-up study. Knee Surg Sports Traumatol Arthrosc 17:1125–1134 4. Baranto A, Hellstro¨m M, Nyman R et al (2006) Back pain and degenerative abnormalities in the spine of young elite divers: a 5-year follow-up magnetic resonance imaging study. Knee Surg Sports Traumatol Arthrosc 14:907–914 5. Botsford DJ, Esses SI, Ogilvie-Harris DJ (1994) In vivo diurnal variation in intervertebral disc volume and morphology. Spine 19:935–940 6. Broberg KB (1993) Slow deformation of intervertebral discs. J Biomech 26:501–512

701 7. Cassidy JJ, Hiltner A, Baer E (1989) Hierarchical structure of the intervertebral disc. Connect Tissue Res 23:75–88 8. Ching CT, Chow DH, Yao FY et al (2004) Changes in nuclear composition following cyclic compression of the intervertebral disc in an in vivo rat-tail model. Med Eng Phys 26:587–594 9. Ekstro¨m L, Kaigle A, Hult E et al (1996) Intervertebral disc response to cyclic loading–an animal model. Proc Inst Mech Eng H 210:249–258 10. Frymoyer JW, Pope MH, Costanza MC et al (1980) Epidemiologic studies of low-back pain. Spine 5:419–423 11. Guo LX, Teo EC (2006) Influence prediction of injury and vibration on adjacent components of spine using finite element methods. J Spinal Disord Tech 19:118–124 12. Hasegawa K, Turner CH, Chen J et al (1995) Effect of disc lesion on microdamage accumulation in lumbar vertebrae under cyclic compression loading. Clin Orthop Relat Res 311:190–198 13. Heliovaara M (1987) Occupation and risk of herniated lumbar intervertebral disc or sciatica leading to hospitalization. J Chronic Dis 40:259–264 14. Holm S, Baranto A, Kaigle Holm A et al (2007) Reactive changes in the adolescent porcine spine with disc degeneration due to endplate injury. Vet Comp Orthop Traumatol 20:12–17 15. Johannessen W, Vresilovic EJ, Wright AC et al (2004) Intervertebral disc mechanics are restored following cyclic loading and unloaded recovery. Ann Biomed Eng 32:70–76 16. Kaigle A, Ekstrom L, Holm S et al (1998) In vivo dynamic stiffness of the porcine lumbar spine exposed to cyclic loading: influence of load and degeneration. J Spinal Disord 11:65–70 17. Kujala UM, Taimela S, Erkintalo M et al (1996) Low-back pain in adolescent athletes. Med Sci Sports Exerc 28:165–170 18. Lundin O, Ekstro¨m L, Hellstro¨m M et al (2000) Exposure of the porcine spine to mechanical compression: differences in injury pattern between adolescents and adults. Eur Spine J 9:466–471 19. Lundin O, Ekstro¨m L, Hellstro¨m M et al (1998) Injuries in the adolescent porcine spine exposed to mechanical compression. Spine 23:2574–2579 20. Lundin O, Hellstro¨m M, Nilsson I et al (2001) Back pain and radiological changes in the thoraco-lumbar spine of athletes. A long-term follow-up. Scand J Med Sci Sports 11:103–109 21. Malko JA, Hutton WC, Fajman WA (1999) An in vivo magnetic resonance imaging study of changes in the volume (and fluid content) of the lumbar intervertebral discs during a simulated diurnal load cycle. Spine 24:1015–1022 22. Natarajan RN, Ke JH, Andersson GB (1994) A model to study the disc degeneration process. Spine 19:259–265 23. Pankoke S, Hofmann J, Wolfel HP (2001) Determination of vibration-related spinal loads by numerical simulation. Clin Biomech (Bristol, Avon) 16(Suppl 1):S45–56 24. Riches PE, Dhillon N, Lotz J et al (2002) The internal mechanics of the intervertebral disc under cyclic loading. J Biomech 35:1263–1271 25. Seenivasan G, Goel V (1994) Applying bone adaptive remodeling theory to ligamentouse spine-preliminary results of partial nucleotomy and stabilization. J Eng Med Biol 13:508–516 26. Swa¨rd L, Hellstro¨m M, Jacobsson B et al (1991) Disc degeneration and associated abnormalities of the spine in elite gymnasts. A magnetic resonance imaging study. Spine 16:437–443 27. Swa¨rd L, Hellstro¨m M, Jacobsson B et al (1990) Back pain and radiologic changes in the thoraco-lumbar spine of athletes. Spine 15:124–129 28. Tsai K, Lin R, Chang G (1998) Rate-related fatigue injury of vertebral disc under axial cyclic loading in a porcine body-disc-body unit. Clin Biomech (Bristol, Avon) 13:S32–S39 29. Wilke HJ, Neef P, Caimi M et al (1999) New in vivo measurements of pressures in the intervertebral disc in daily life. Spine 24:755–762

123