Characterization of articular cartilage by combining microscopic analysis with a fibril-reinforced finite-element model

https://doi.org/10.1016/j.jbiomech.2006.07.026Get rights and content

Abstract

Load-bearing characteristics of articular cartilage are impaired during tissue degeneration. Quantitative microscopy enables in vitro investigation of cartilage structure but determination of tissue functional properties necessitates experimental mechanical testing. The fibril-reinforced poroviscoelastic (FRPVE) model has been used successfully for estimation of cartilage mechanical properties. The model includes realistic collagen network architecture, as shown by microscopic imaging techniques. The aim of the present study was to investigate the relationships between the cartilage proteoglycan (PG) and collagen content as assessed by quantitative microscopic findings, and model-based mechanical parameters of the tissue. Site-specific variation of the collagen network moduli, PG matrix modulus and permeability was analyzed. Cylindrical cartilage samples (n=22) were harvested from various sites of the bovine knee and shoulder joints. Collagen orientation, as quantitated by polarized light microscopy, was incorporated into the finite-element model. Stepwise stress-relaxation experiments in unconfined compression were conducted for the samples, and sample-specific models were fitted to the experimental data in order to determine values of the model parameters. For comparison, Fourier transform infrared imaging and digital densitometry were used for the determination of collagen and PG content in the same samples, respectively. The initial and strain-dependent fibril network moduli as well as the initial permeability correlated significantly with the tissue collagen content. The equilibrium Young's modulus of the nonfibrillar matrix and the strain dependency of permeability were significantly associated with the tissue PG content. The present study demonstrates that modern quantitative microscopic methods in combination with the FRPVE model are feasible methods to characterize the structure–function relationships of articular cartilage.

Introduction

Articular cartilage consists of a three-dimensional collagen network, proteoglycans (PGs), interstitial water and chondrocytes (Buckwalter and Martin, 1995; Mow et al., 1991). Several external and internal factors, e.g. abnormal or impact mechanical loading or joint inflammation, may lead to degradation of collagen and PGs with a consequence of impaired functional properties of the tissue (Armstrong and Mow, 1982; Mow et al., 1992). Cartilage injury may further lead to osteoarthritis (OA) (Buckwalter and Martin, 1995). Structural changes in the cartilage matrix can be revealed by imaging techniques, such as high resolution magnetic resonance imaging (MRI) or quantitative microscopy (Park et al., 2003; Rieppo et al., 2003; Wheaton et al., 2005). However, these imaging techniques are unable to quantitate changes in functional properties of the cartilage, sensitively impaired in early OA (Saarakkala et al., 2003). Those changes can only be reliably revealed by experimental mechanical tests. However, optimal interpretation of the mechanical data necessitates the use of a valid theoretical model (Guilak et al., 1994; Korhonen et al., 2003; Mow et al., 1991; Saarakkala et al., 2004).

Several theoretical models have been developed to characterize mechanical behavior of articular cartilage. Most of them cannot capture the effect of depth-dependent collagen structure on the mechanical characteristics (Cohen et al., 1998; DiSilvestro et al., 2001; Li et al., 1999; Soltz and Ateshian, 2000). Li et al. (2000) modeled the depth-wise mechanical behavior of collagen, however, actual collagen network structure was not included. Inhomogeneous and anisotropic structure of articular cartilage can be retrieved from MRI or microscopical analyses (Király et al., 1997; Nieminen et al., 2001). By taking into account the structural inhomogeneity and depth dependency, determination of the early stages of cartilage degeneration may be more realistic. In addition, prediction of the onset and progression of OA, e.g. due to abnormal mechanical loading after meniscectomy, may be improved (LeRoux et al., 2000; Wilson et al., 2003, Wilson et al., 2006). Inherently, stress and strain distributions will change in cartilage during tissue degeneration, inducing severe variations e.g. in the mechanical signals sensed by the chondrocytes (Guilak et al., 1994; Mow et al., 1992). Alterations in the depth-dependent collagen architecture may play a critical role in these pathologies. The fibril-reinforced poroviscoelastic (FRPVE) model (Wilson et al., 2004) includes depth-dependent collagen orientation and takes into account the compression tension nonlinearity of articular cartilage. Thereby, the microscopical FRPVE model may be used to characterize local structure–function relationships and interstitial stress/strain patterns of cartilage.

Earlier, structure–function relationships in articular cartilage have been studied by combining structural analysis and mechanical tests (LeRoux et al., 2000; Nieminen et al., 2000). Most of these studies have not utilized theoretical models in order to obtain more comprehensive relationships between the structure and mechanical behavior of collagen network, PGs and fluid. A few studies have combined theoretical models with the structural analysis of collagen or PGs (Lu et al., 2004; Tran-Khanh et al., 2005), but they lack the realistic anisotropic collagen orientation or strain-dependent permeability of cartilage and do not consider site-dependent changes in the mechanical and structural properties of the tissue.

In the present study, the main hypothesis was that the experimentally determined collagen and PG contents can be related to corresponding model components, i.e. fibril network stiffness and PG matrix stiffness, respectively. These relations have not been addressed simultaneously before. Thereby, we aimed to investigate the relationships between the FRPVE model parameters of bovine articular cartilage and the tissue collagen and PG content, as assessed experimentally using quantitative microscopy. For this purpose, mechanical testing, microscopic analyses and finite-element (FE) analyses were carried out for cartilage samples from the bovine knee and shoulder joints. Topographical variations in the values of model parameters, as derived from the analyses of experimental stress-relaxation measurements, and predictions for stress and lateral displacement distributions are also presented.

Section snippets

Sample preparation

Bovine knee and shoulder joints were supplied by a local slaughterhouse (Atria Oy, Kuopio, Finland). Cylindrical, full thickness cartilage samples from bovine femur, tibia, patella and humerus were harvested using a razor blade and a biopsy punch of 3.7 mm in diameter (Korhonen et al., 2002). Before mechanical testing, the sample diameters were measured with a stereomicroscope (Nikon SMZ-10, Nikon Inc., Japan) (Table 1).

Mechanical testing

Stepwise stress-relaxation experiments in unconfined compression geometry

Results

The experimental stress-relaxation tests could be successfully simulated, as shown by the accurately fitted model curves to the experimental data (Fig. 2). E0 and Eε correlated linearly with the collagen content (r=0.74, p<0.001 and r=0.82, p<0.001, respectively, Fig. 3, Table 2), as measured with FTIRI. Lower, but significant linear associations were found between the fibril network moduli (E0 and Eε) and PG content (OD) (r=0.54, p<0.01 and r=0.55, p<0.01, respectively, Table 2). Em and M

Discussion

In the present study, structure–function relationships of bovine articular cartilage were characterized by combining information from the quantitative microscopy with the depth-dependent FE analyses. Model parameters correlated significantly with the collagen and PG contents. All structural and mechanical parameters showed site-specific variations in the bovine knee and humeral cartilage. Concomitantly, stress and strain distributions, as characterized by the FRPVE model, showed specific

Acknowledgments

Financial support from the Kuopio University Hospital, Kuopio, Finland (EVO 5031335); The Academy of Finland (Project nos. 206255 and 205886); The Alberta Heritage Foundation for Medical Research, Alberta, Canada; Sigrid Jusélius Foundation, Helsinki, Finland; The Invalid Foundation, Helsinki, Finland; The Finnish Cultural Foundation of Northern Savo, Kuopio, Finland; Foundation for Technology Advancement, Helsinki, Finland; and Biomaterial Graduate School, Helsinki, Finland is acknowledged.

References (53)

  • S. Saarakkala et al.

    Ultrasound indentation of normal and spontaneously degenerated bovine articular cartilage

    Osteoarthritis and Cartilage

    (2003)
  • N. Tran-Khanh et al.

    Aged bovine chondrocytes display a diminished capacity to produce a collagen-rich, mechanically functional cartilage extracellular matrix

    Journal of Orthopaedic Research

    (2005)
  • J.L. Wang et al.

    Failure criterion of collagen fiber: viscoelastic behavior simulated by using load control data

    Theoretical and Applied Fracture Mechanics

    (1997)
  • X. Wang et al.

    Age-related changes in the collagen network and toughness of bone

    Bone

    (2002)
  • A.J. Wheaton et al.

    Detection of changes in articular cartilage proteoglycan by T1p magnetic resonance imaging

    Journal of Orthopaedic Research

    (2005)
  • W. Wilson et al.

    Pathways of load-induced cartilage damage causing cartilage degeneration in the knee after meniscectomy

    Journal of Biomechanics

    (2003)
  • W. Wilson et al.

    Stresses in the local collagen network of articular cartilage: a poroviscoelastic fibril-reinforced element study

    Journal of Biomechanics

    (2004)
  • W. Wilson et al.

    A fibril-reinforced poroviscoelastic swelling model for articular cartilage

    Journal of Biomechanics

    (2005)
  • W. Wilson et al.

    Erratum to Stresses in the local collagen network of articular cartilage: a poroviscoelastic fibril-reinforced finite element study [2004. Journal of Biomechanics 37, 357–366] and A fibril-reinforced poroviscoelastic swelling model for articular cartilage [2005. Journal of Biomechanics 38, 1195–1204]

    Journal of Biomechanics

    (2005)
  • M. Wong et al.

    Volumetric changes of articular cartilage during stress relaxation in unconfined compression

    Journal of Biomechanics

    (2000)
  • C.G. Armstrong et al.

    Variations in the intrinsic mechanical properties of human articular cartilage with age, degeneration, and water content

    The Journal of Bone and Joint Surgery, American Volume

    (1982)
  • J.P. Arokoski et al.

    Decreased birefringence of the superficial zone collagen network in the canine knee (stifle) articular cartilage after long distance running training, detected by quantitative polarised light microscopy

    Annals of the Rheumatic Diseases

    (1996)
  • A. Benninghoff

    Form und Bau der Gelenkknorpel in ihren Beziehungen zur function

    Zeitschrift für Zellforschung

    (1925)
  • J.A. Buckwalter et al.

    Degenerative joint disease

    Clinical Symposia

    (1995)
  • B. Cohen et al.

    A transversely isotropic biphasic model for unconfined compression of growth plate and chondroepiphysis

    Journal of Biomechanical Engineering

    (1998)
  • M.R. DiSilvestro et al.

    Biphasic poroviscoelastic simulation of the unconfined compression of articular cartilage: I—simultaneous prediction of reaction force and lateral displacement

    Journal of Biomechanical Engineering

    (2001)
  • Cited by (0)

    View full text