International Cartilage Repair Society

Transcription

1 OsteoArthritis and Cartilage (2005) 13, 964e970 ª 2005 OsteoArthritis Research Society International. Published by Elsevier Ltd. All rights reserved. doi: /j.joca Prestrain decreases cartilage susceptibility to injury by ramp compression in vitro V. Morel Ph.D., A. Merçay M.Sc. and T. M. Quinn Ph.D.* Cartilage Biomechanics Group, Ecole Polytechnique Fédérale de Lausanne (EPFL), Switzerland Summary International Cartilage Repair Society Background: Injurious mechanical loading of articular cartilage can be an initiating factor in the development of degenerative joint disease. The tissue response to compression depends on the loading conditions and matrix mechanical properties. The short-term loading history of cartilage can affect its water content and microstructural organization, and may thereby modify its susceptibility to injury. We therefore examined the role of prestrain on the response of articular cartilage to injurious compression. Methods: The full-thickness cartilage of bovine osteochondral explants was subjected to prestrains of 0, 5, 10, 25 or 50% before application of injurious ramp compression characterized by a strain rate of 7! 10 ÿ2 or 7! 10 ÿ3 s ÿ1 and peak stress of 3.5 or 14 MPa. Effects of prestrain were evaluated in terms of fluid exudation, tissue mechanical stiffening, and the tissue response to injurious compression as characterized by macroscopic crack formation, cell viability and glycosaminoglycan release to culture media. Results: Macroscopic crack formation due to injurious compression decreased with increasing prestrain in association with lower cell mortality. Significantly decreased susceptibility to injury was already evident for 10% prestrain. In contrast, explant mechanical stiffness was unchanged up to 25% prestrain. Conclusion: Findings demonstrate that compressive strains due to the short-term loading history of cartilage may strongly reduce its susceptibility to mechanical injury. Conversely, matrix swelling may render cartilage more vulnerable to injury. The cartilage response to injurious compression is therefore strongly influenced by matrix fluid content, and possibly also by other structural parameters such as collagen fiber orientation. ª 2005 OsteoArthritis Research Society International. Published by Elsevier Ltd. All rights reserved. Key words: Cartilage, Injury, Compression, Strain, Strain rate. Introduction Nonphysiological mechanical loading is implicated in the initiation and progression of cartilage degeneration. Mechanical loading conditions ranging from very high strain rate impact loading conditions (as in traumatic accidents) to very low strain rate loading (as in obesity or joint misalignment) may all contribute to cartilage injury and degenerative joint disease. Traumatic joint injury by blunt mechanical insult has been demonstrated to be a risk factor for the development of osteoarthritis 1. In vitro studies of impact loading 2,3 and high strain rate injurious compression 4e7 of cartilage have shown that cell death and matrix damage resulting from fluid pressurization and collagen network tensile failure 8 could represent initiating events for degenerative processes. Therefore, acute mechanical events can have important long-term implications for cartilage disorders. Cyclic loading can also lead to chondrocyte death 9e11 or mechanical weakening of cartilage 12, with effects that worsen over an extended period of applied loads. These studies may represent simulations of joint overuse (as opposed to more abrupt traumatic injury) and emphasize the contribution of cartilage short-term loading *Address correspondence and reprint requests to: Thomas M. Quinn, Cartilage Biomechanics Group, AAB019, EPFL Station 15, CH-1015 Lausanne, Switzerland. Tel: ; Fax: ; thomas.quinn@epfl.ch Received 17 March 2004; revision accepted 21 June history in injurious processes. It is reasonable to hypothesize that short-term loading history may also contribute to changing risks of tissue injuries throughout the course of sporting events 13. Mechanical properties are conferred on articular cartilage by the composition and structure of its extracellular matrix (ECM); therefore modifications at the matrix level could affect the tissue response to injurious compression. Alteration of the composition and structure of the ECM can occur in the context of degenerative joint disease or trauma 14,15, after surgery such as graft implantation 16, in the absence of locomotion 17 or when joint loading is accentuated 18. These alterations can modify the mechanical properties of the tissue and may change its local susceptibility to injury. Mechanical loading is also commonly used to improve the development of engineered cartilage in vitro 19, but the response of the growing tissue to applied load may vary as intratissue mechanical properties evolve. Cartilage matrix density, microstructural organization, and mechanical properties can also change due to mechanical loading in the absence of cell-mediated remodeling, simply due to changes in water content. Therefore, better understanding of the effects of short-term loading history on the cartilage response to injurious compression may have important implications for the prevention of degenerative joint disease and also for optimization of tissue engineering methods. We hypothesized that the short-term loading history of cartilage can influence its subsequent response to injurious compression due to modifications in tissue structural 964

2 Osteoarthritis and Cartilage Vol. 13, No organization and mechanical properties. Therefore, prestrains of 0, 5, 10, 25 and 50% were applied to cartilage in osteochondral explants and followed by injurious compression characterized by strain rates of 7! 10 ÿ2 or 7! 10 ÿ3 s ÿ1 and peak stresses of 3.5 and 14 MPa. Effects of prestrain were evaluated in terms of cartilage fluid content and equilibrium and dynamic moduli, and compared with the tissue response to injurious compression quantified in terms of macroscopic superficial cracks, cell viability, glycosaminoglycan (GAG) release to media, and changes in wet weight. Material and methods DISSECTION AND TISSUE CULTURE Osteochondral cores (4 mm diameter) were obtained from fresh 18-month-old bovine humeral heads using a biopsy drill (Straumann Medical) under saline irrigation. Bone was trimmed to 1 mm thickness with a vibrating saw (Stratec Medical) and the full-thickness cartilage layer to 2.7 mm diameter using a dermal punch (Stiefel Laboratorium) and scalpel. Explants were incubated (at 37(C and 5% CO 2 )in dulbecco s modified eagle medium (Oxoid) containing 1 g/l D-Glucose, 3.7 g/l NaHCO 3, 1 g/l N-acetyl-lanalyl-L-glutamine, and supplemented with 0.1 mm non-essential amino acids (Oxoid), 0.4 mml-proline (Sigma), 10 mg/ml streptomycin sulfate, 10,000 units/ml penicillin G sodium, 25 mg/ ml amphotericin B (Gibco), and 10% (v/v) fetal bovine serum (Sigma), with daily changes of 1 ml media per explant for 10 days. MECHANICAL LOADING After 5 days of culture, explants were compressed axially without radial confinement between a steel loading post on the articular surface, and a steel and plexiglass support chamber. A displacement actuator (PM500-1A, Newport) raised the support chamber to push the explant against the loading post which was attached to a load cell (Model 31, Sensotec). The actuator and load cell were fixed in an aluminum and stainless steel frame, and interfaced with a microcomputer (Macintosh) with instrumentation software (LabVIEW) continuously recording stress and strain vs time. Stress was defined relative to explant articular surface area at dissection, and strain was defined relative to the cartilage thickness measured under a dissection microscope (NCL 150, Nikon AG) just before loading. Throughout loading, explants were immersed in culture media but exposed to room temperature and atmospheric conditions. The loading protocol was applied in three successive steps [Fig. 1(a)]. First, a prestrain of 0, 5, 10, 25 or 50% was established by ramp compression at a strain rate of 7! 10 ÿ4 s ÿ1 (which was previously found to induce no injury to cartilage 8 ). Then, explants were held at the fixed prestrain for a 15-min relaxation period, during which a stable equilibrium load was attained. Finally a single ramp compression characterized by a strain rate of 7! 10 ÿ2 or 7! 10 ÿ3 s ÿ1 to a maximal stress of 3.5 or 14 MPa was applied. Strain rates were normalized to the non-injurious rate of 7! 10 ÿ4 s ÿ1 for ease of presentation. The last part of the loading protocol (normalized strain rate _3 N Z10 or 100) was designated injurious compression. Previous work has shown that these loading conditions (without prestrain) result in cartilage injury of moderate severity 8 ; they were therefore used in the present study to investigate the influence of prestrain on mechanical injury (for better or for worse). Four Fig. 1. (a) The loading protocol consisted of a non-injurious ramp prestrain to 0, 5, 10, 25 or 50%, mechanical relaxation to load equilibrium for 15 min, and a single injurious ramp compression. (b) Mechanical parameters determined from measured stressestrain relationships included equilibrium modulus (E eq Zs eq /3 eq ), dynamic moduli (E dyn ) near stresses of 3.5 and 14 MPa, and maximum strain (3 max ). parameters were determined from stressestrain curves [Fig. 1(b)]: equilibrium modulus E eq, two dynamic moduli E dyn, and maximal strain 3 max. E eq was defined as the ratio of the equilibrium stress s eq at the end of the relaxation period and the prestrain 3 eq. E dyn was defined as the slope of a least-squares fit to stress vs strain during injurious compression for stresses in the ranges 2.5e3.5 MPa and 12e14 MPa. Explants were weighed on an analytical balance (Sarto BP80) immediately before and after the loading protocol (after removal of surface fluid with a tissue) to determine changes in water content due to loading. Control explants were subjected to prestrain and relaxation only (prestrained controls) or to no loading at all (uncompressed controls). HISTOLOGICAL AND BIOCHEMICAL ANALYSIS Four days after the loading protocol, explants were removed from culture, weighed again, examined with the naked eye for the presence of macroscopic cracks, and assessed histologically for chondrocyte viability. As previously described 8, w1 mm thick vertical slices of cartilage were obtained manually from the middle of explants with a razor blade and soaked for 5 min at room temperature in saline containing 2 mm calcein AM and 4 mm ethidium homodimer-1. Confocal fluorescence imaging (LSM 410, Carl Zeiss AG) using 495 nm excitation and separation of

3 966 V. Morel et al.: Prestrain decreases cartilage susceptibility to injury green (515 nm) and red (635 nm) fluorescence was used to visualize spatial distributions of live and dead cells. A 10! objective was used; six images (1.28! 1.28 mm) were combined to obtain one of the entire explant slice. A human user manually traced areas of images containing cartilage (regardless of cell viability) and containing viable cells with a normal appearance. The boundary between live and dead cells was almost always clearly defined. These identified image surface areas were then output to analysis software (Matlab, The Mathworks Inc.) and used to estimate the proportion of tissue containing viable cells. Explants were then digested in 1 ml of 25 mg/ml papain (Sigma) in saline at 60(C overnight. Explant digests and spent media were assayed for GAG content 20. STATISTICAL ANALYSIS Before the loading protocol, each explant was randomly assigned to an experimental group characterized by level of prestrain and injurious compression strain rate and maximum stress. Dissections from 14 different animals contributed explants to this study. Significant differences (P! 0.05) among prestrain-related effects were identified by analysis of variance (ANOVA) with post hoc Tukey tests, and differences between injured and control explants were identified by ANOVA with post hoc Dunett tests. Data are reported as mean G S.E.M. (n). Results MECHANICS OF LOADING Equilibrium stress increased approximately linearly with prestrain from 0 to 25%, then increased more dramatically for 50% prestrain. This corresponded to a relatively constant equilibrium elastic modulus (E eq ) of approximately 0.27 MPa for 5, 10 and 25% prestrain and an increased E eq of G MPa (n Z 39) at 50% prestrain [Fig. 2(a)]. Normalized water loss in prestrained controls (mass of exuded water divided by 1 g/ml and divided by initial cartilage volume measured under a dissection microscope) increased from G (n Z 10) at 5% prestrain to G (n Z 9) at 50% prestrain [Fig. 2(b)]. Consistent with these findings, the additional water loss associated with injurious compression tended to decrease with increasing prestrain [Fig. 2(b)]. The stiffness and general mechanical behavior of explants were not strongly affected by prestrain up to 25%. During injurious compression, the slopes of stressestrain curves continuously increased [Fig. 3(a)]. At a normalized strain rate (_3 N ) of 10, the first part of the curve was flatter than at _3 N Z100. As a consequence, larger strains were attained for the same stress at the lower strain rate. However, E dyn at 3.5 and 14 MPa was not significantly different for_3 N Z10 or 100 [Fig. 3(b)]. At both injurious strain rates, for stresses from w3 to 14 MPa, stressestrain curves were almost superimposed for all prestrains %25%, and therefore E dyn was not different among prestrains in this range. After 50% prestrain and for both strain rates, the stressestrain relationship became more linear. E dyn did not increase dramatically (between 3.5 and 14 MPa, for 50% prestrain) during deformation, but was always greater than for lesser prestrains [Fig. 3(b)]. CELL AND MATRIX RESPONSES TO COMPRESSION Uncompressed controls always exhibited viable cells throughout all tissue zones and an intact articular surface Fig. 2. (a) Equilibrium moduli at the end of the relaxation period following prestrain (n Z 39e62). +: P! 0.05; lines indicate data between which significant differences were observed. (b) Normalized water loss in controls subjected to prestrain plus relaxation (white) and during the full loading protocol including injurious compression at _3 N Z10 (gray) and 100 (black) to 14 MPa (n Z 8e13). [Fig. 4(a)], while injurious compression introduced cracks in the articular surface of some explants and contributed to decreased cell viability near the superficial zone [Fig. 4(bed)]. Prestrain reduced the occurrence of macroscopic cracks in the articular surface (which is the only place where cracks were observed). After 5% prestrain, 58.3% (n Z 12) of explants were fissured when compressed to 14 MPa at _3 N Z100 strain rate [Fig. 4(c)], compared to 83.3% (n Z 12) of explants subjected to the same injurious compression but without prestrain [Fig. 4(b)]. No explants were fissured after 50% prestrain (n Z 9). The same trend of diminishing cracks with increasing prestrain was observed for compression at _3 N Z10: without prestrain, 27.3% (n Z 11) of explants were fissured but no fissures were found for prestrains R 25% (n Z 10) [Fig. 5(a)]. No macroscopic superficial cracks were found on explants compressed to 3.5 MPa nor in prestrained controls. The smallest proportion of tissue vertical section surface area containing viable cells on confocal images was found for explants subjected to no prestrain and injurious compression to 14 MPa at _3 N Z100, with a value of 71.1 G 6.6% (n Z 7) [Figs. 4(b) and 5(b)]. Uncompressed controls exhibited some dead cells in the superficial zone and near radial edges of explant disks, indicating limitations of our tissue source and dissection methods. For all other loading

4 Osteoarthritis and Cartilage Vol. 13, No Fig. 3. (a) Mean stressestrain relationships (n Z 8e12) for injurious compression after no prestrain (white), 25% prestrain (gray) and 50% prestrain (black). Squares and circles indicate compression at normalized strain rates (_3 N ) of 10 and 100, respectively. (b) Dynamic moduli determined from the mean slope of stressestrain curves between 12 and 14 MPa (n Z 8e13). Prestrain (%) is indicated on the bars. +: P! 0.05; lines indicate data between which significant differences were observed. conditions, including prestrained controls, the area fraction containing viable cells on confocal images was between 80 and 90%, with no significant difference among these groups. For injurious compression at _3 N Z100 to 14 MPa, viability without prestrain was significantly less than after 10, 25 and 50% prestrain [Figs. 4(b,d) and 5(b)]. Prestrain had no significant effect on GAG release to media (normalized to explant GAG contents) after injurious compression (not shown). Elevated GAG release was observed the day after explant dissection followed by stabilization at a lower level for the next days, with no significant effects of loading protocols. For all conditions, wet weight 4 days after the loading protocol was not significantly different than wet weight before the loading protocol (not shown). Discussion Prestrain of cartilage explants prior to injurious loading reduced the occurrence of superficial cracks and associated cell death. These effects were surprisingly abrupt insofar as significant reductions in injury were evident with the application of only 10% prestrain, compared to explants subjected to injurious compression without prestrain. Fig. 4. Confocal microscopy images of cell viability within the cartilage layer of osteochondral explants. The articular surface is at the top of each image and the tidemark at the bottom. (a) Uncompressed controls exhibited viable cells throughout all tissue zones and an intact articular surface. Injurious compression at _3 N Z100 to 14 MPa after (b) 0% and (c) 5% prestrain often resulted in cracks in the superficial zone (white arrows) and associated decreases in cell viability, (d) 10% prestrain appeared to decrease the severity of cell and matrix injury for the same injurious compression parameters. Despite these biologically important changes, tissue mechanical behavior (equilibrium and dynamic moduli, stress response to applied strain) was not significantly affected by prestrains up to 25%. This contrast highlights the interactions between phenomena at tissue, cell, and molecular length scales which contribute to cartilage mechanical injury.

5 968 V. Morel et al.: Prestrain decreases cartilage susceptibility to injury Fig. 5. (a) Fraction of explants exhibiting macroscopic superficial cracks after injurious compression at normalized strain rates (_3 N )of 10 (gray) and 100 (black) to 14 MPa (n Z 9e12). (b) Cell viability in uncompressed controls (white) and after injurious compression at _3 N Z100 to 14 MPa (black), as quantified by area fraction containing viable cells on confocal images (n Z 6e7). +: P! 0.05; lines indicate data between which significant differences were observed. As previously observed 5,8, cartilage injury under relatively high strain rate (_3 N O1) ramp compression involved cell death primarily near the superficial zone and in association with macroscopic cracks in the articular surface (Fig. 4). Injurious loading induced acute changes in wet weight [Fig. 2(b)], but neither wet weight nor GAG release was different from uncompressed controls 4 days after injury. Injury was most severe at the highest strain rate and for the highest peak stress, also consistent with previous studies involving impact loading of cartilage 5,7,8. Prestrain dramatically reduced the severity of injuries due to subsequent injurious compression (Figs. 4 and 5). However, none of these manifestations of cell and matrix injury in response to ramp compression was changed by prestrain. Therefore, the effects of prestrain were to alter the susceptibility of cartilage to mechanical injury by ramp compression, rather than to change the nature of injury when it did occur. Tissue-scale mechanics of cartilage explants, in terms of static [Fig. 2(a)] and dynamic elastic moduli [Fig. 3(b)] and stressestrain behavior during injurious compression [Fig. 3(a)], were very constant up to at least 25% prestrain. This stability of mechanical behavior may reflect a physiological requirement of consistent tissue function despite a wide range of possible short-term loading histories. Concomitantly, however, important changes in matrix water content [Fig. 2(b)] and matrix density were clear, indicating that matrix mechanical properties such as hydraulic permeability k and bulk modulus H A were likely affected by prestrain 21. Specifically, increased matrix GAG density was likely associated with an increase in H A but a decrease in k. The increase in H A would have had relatively small effects on the tissue-scale equilibrium modulus due to the interplay between GAG swelling pressure, collagen tension 22,23, and the complementary roles of radially and axially oriented collagen fibers in unconfined compression of cartilage 24,25. Furthermore, decreases in k and increases in H A may have offset each other for prestrains up to 25% since the poroelastic mechanics of cartilage are governed largely by the product H A k in this range 26, rather than by H A or k individually. At the extreme of 50% prestrain, explants were clearly more rigid [Figs. 2(a) and 3(b)] and more elastic as opposed to poroelastic insofar as they exhibited stress vs strain that varied far less with strain rate [Fig. 3(a)]. Apparently, the severe reduction in matrix water content [Fig. 2(b)] at 50% prestrain had fundamentally altered their mechanical response to injurious compression. Prestrain strongly reduced the occurrence of macroscopically visible cracks in the articular surface. Previously, cracks were found to be closely associated with matrix fluid pressurization during explant axial compression and were hypothesized to result from collagen network tensile overloading, particularly for collagen fibers oriented in the explant radial direction during unconfined compression 8,25. In the present study, prestrain involved exudation of matrix fluid under relatively low pressures, contributing to less fluid outflow during subsequent injurious compression [Fig. 2(b)]. At the end of the relaxation period [Fig. 1(a)], the maximum fluid volume had been evacuated for each predetermined level of axial strain. In addition, it is possible that mechanical relaxation allowed for more complete reorientation of collagen fibers towards the explant radial direction under axial compression 27. Therefore, the matrix was as mechanically relaxed as possible for the applied level of prestrain, and better able to accommodate fluid pressurization and radial tensile loading during injurious compression. Tissue prestrains applied at a relatively slow rate may have therefore resulted in decreased matrix damage by reducing the molecular-scale tensile strains arising during high strain rate injury. Modeling approaches which address relationships between tissue-scale mechanics and matrix molecular physics may help to clarify these phenomena 27. Cell death was reduced with the application of prestrain, concomitantly with the reduction of matrix cracks. Lower cell mortality was likely a direct effect of the reduction of matrix cracks by prestrain. Under high strain rate injurious loading, acute cell death is thought to result from direct trauma due to solid deformations and fluid shear stresses associated with matrix disruption 5,6. The association between matrix cracks and cell death is consistent across a range of studies 5,7,8, but it remains somewhat unclear why cracks and cell death are often localized primarily to superficial tissue zones. This trend suggests that the mechanical behavior of the superficial zone, and its interactions with other tissue zones during injurious loading play important roles in determining its susceptibility to injury 24. Furthermore, cell-scale mechanical events particular to the superficial zone may play important roles in mediating the cell response to tissue injurious loading. A central theme in the present and previous studies appears to be the role of fluid content within cartilage, under

6 Osteoarthritis and Cartilage Vol. 13, No. 11 a wide range of loading conditions, as a contributing factor in tissue mechanical injury. After removal from a joint surface, cartilage swells by 3e15% 28. Present results demonstrate that a counteracting level of prestrain significantly reduces the negative effects of injurious compression, suggesting that increases in matrix water content above that of the normal in situ state can render cartilage more vulnerable to mechanical injury. Cartilage swelling is one of the earliest manifestations of osteoarthritis 29. Present results imply that this alone may render cartilage more vulnerable to mechanical injury in vivo, and facilitate mechanically-mediated progression of degradative disorders. Swelling may also occur due to tissue slicing during surgical procedures, and may therefore represent a potentially negative side-effect. Present results may also partially explain the efficacy of postoperative continuous passive motion 30, insofar as joint loading may help to maintain normal cartilage water content and prevent localized swelling, thereby decreasing cartilage susceptibility to mechanical injury. Results may therefore be relevant in a wide range of clinical contexts, including injury prevention and treatment, and in the optimization of surgical or tissue engineering methods. Acknowledgments This work was supported by a grant from the AO Foundation, Switzerland. References 1. Wilder FV, Hall BJ, Barrett JP Jr, Lemrow NB. History of acute knee injury and osteoarthritis of the knee: a prospective epidemiological assessment. The Clearwater Osteoarthritis Study. Osteoarthritis Cartilage 2002;10(8):611e6. 2. Borrelli J Jr, Torzilli PA, Grigiene R, Helfet DL. Effect of impact load on articular cartilage: development of an intra-articular fracture model. J Orthop Trauma 1997; 11(5):319e Atkinson TS, Haut RC, Altiero NJ. Impact-induced fissuring of articular cartilage: an investigation of failure criteria. J Biomech Eng 1998;120(2):181e7. 4. Quinn TM, Grodzinsky AJ, Hunziker EB, Sandy JD. Effects of injurious compression on matrix turnover around individual cells in calf articular cartilage explants. J Orthop Res 1998;16(4):490e9. 5. Ewers BJ, Dvoracek-Driksna D, Orth MW, Haut RC. The extent of matrix damage and chondrocyte death in mechanically traumatized articular cartilage explants depends on rate of loading. J Orthop Res 2001;19(5): 779e Kurz B, Jin M, Patwari P, Cheng DM, Lark MW, Grodzinsky AJ. Biosynthetic response and mechanical properties of articular cartilage after injurious compression. J Orthop Res 2001;19(6):1140e6. 7. Quinn TM, Allen RG, Schalet BJ, Perumbuli P, Hunziker EB. Matrix and cell injury due to sub-impact loading of adult bovine articular cartilage explants: effects of strain rate and peak stress. J Orthop Res 2001;19(2):242e9. 8. Morel V, Quinn TM. Cartilage injury by ramp compression near the gel diffusion rate. J Orthop Res 2004; 22(1):145e Levin A, Burton-Wurster N, Chen CT, Lust G. Intercellular signaling as a cause of cell death in cyclically impacted cartilage explants. Osteoarthritis Cartilage 2001;9(8):702e Clements KM, Bee ZC, Crossingham GV, Adams MA, Sharif M. How severe must repetitive loading be to kill chondrocytes in articular cartilage? Osteoarthritis Cartilage 2001;9(5):499e Chen CT, Bhargava M, Lin PM, Torzilli PA. Time, stress, and location dependent chondrocyte death and collagen damage in cyclically loaded articular cartilage. J Orthop Res 2003;21(5):888e Thibault M, Poole AR, Buschmann MD. Cyclic compression of cartilage/bone explants in vitro leads to physical weakening, mechanical breakdown of collagen and release of matrix fragments. J Orthop Res 2002;20(6):1265e Bixler B, Jones RL. High-school football injuries: effects of a post-halftime warm-up and stretching routine. Fam Pract Res J 1992;12(2):131e Ewers BJ, Newberry WN, Haut RC. Chronic softening of cartilage without thickening of underlying bone in a joint trauma model. J Biomech 2000;33(12): 1689e Kaab MJ, Ito K, Clark JM, Notzli HP. The acute structural changes of loaded articular cartilage following meniscectomy or ACL-transection. Osteoarthritis Cartilage 2000;8(6):464e Ahsan T, Sah RL. Biomechanics of integrative cartilage repair. Osteoarthritis Cartilage 1999;7(1):29e Jurvelin J, Kiviranta I, Tammi M, Helminen JH. Softening of canine articular cartilage after immobilization of the knee joint. Clin Orthop 1986;207: 246e Jurvelin J, Kiviranta I, Saamanen AM, Tammi M, Helminen HJ. Indentation stiffness of young canine knee articular cartilage-influence of strenuous joint loading. J Biomech 1990;23(12):1239e Demarteau O, Jakob M, Schafer D, Heberer M, Martin I. Development and validation of a bioreactor for physical stimulation of engineered cartilage. Biorheology 2003; 40(1e3):331e Farndale RW, Sayers CA, Barrett AJ. A direct spectrophotometric microassay for sulfated glycosaminoglycans in cartilage cultures. Connect Tissue Res 1982;9(4):247e Grodzinsky AJ. Electromechanical and physicochemical properties of connective tissues. CRC Crit Rev Biomed Eng 1983;9(2):133e Maroudas A. Balance between swelling pressure and collagen tension in normal and degenerate cartilage. Nature 1976;260:808e Park S, Krishnan R, Nicoll SB, Ateshian GA. Cartilage interstitial fluid load support in unconfined compression. J Biomech 2003;36(12):1785e Flachsmann R, Broom ND, Hardy AE. Deformation and rupture of the articular surface under dynamic and static compression. J Orthop Res 2001;19(6): 1131e Morel V, Quinn TM. Short-term changes in cell and matrix damage following mechanical injury of articular cartilage explants and modelling of microphysical mediators. Biorheology 2004;41(3e4):509e Mow VC, Kuei SC, Lai WM, Armstrong CG. Biphasic creep and stress relaxation of articular cartilage in compression: theory and experiments. J Biomech Eng 1980;102(l):73e84.

7 970 V. Morel et al.: Prestrain decreases cartilage susceptibility to injury 27. Farquhar T, Dawson PR, Torzilli PA. A microstructural model for the anisotropic drained stiffness of articular cartilage. J Biomech Eng 1990;112:414e Setton LA, Tohyama H, Mow VC. Swelling and curling behaviors of articular cartilage. J Biomech Eng 1998; 120(3):355e Buckwalter JA, Mankin HJ. Articular cartilage: degeneration and osteoarthritis, repair, regeneration, and transplantation. Instr Course Lect 1998;47: 487e Alfredson H, Lorentzon R. Superior results with continuous passive motion compared to active motion after periosteal transplantation. A retrospective study of human patella cartilage defect treatment. Knee Surg Sports Traumatol Arthrosc 1999;7(4): 232e8.