In vitro determination of biomechanical properties of human articular cartilage in osteoarthritis using multi-parametric MRI

Abstract

The objective of this study was to evaluate the correlations between MR parameters and the biomechanical properties of naturally degenerated human articular cartilage. Human cartilage explants from the femoral condyles of patients who underwent total knee replacement were evaluated on a micro-imaging system at 3 T. To quantify glycosaminoglycan (GAG) content, delayed gadolinium-enhanced MRI of the cartilage (dGEMRIC) was used. T₂ maps were created by using multi-echo, multi-slice spin echo sequences with six echoes: 15, 30, 45, 60, 75, and 90 ms. Data for apparent diffusion constant (ADC) maps were obtained from pulsed gradient spin echo (PGSE) sequences with five b-values: 10.472, 220.0, 627.0, 452.8, 724.5, and 957.7. MR parameters were correlated with mechanical parameters (instantaneous (I) and equilibrium (Eq) modulus and relaxation time (τ)), and the OA stage of each cartilage specimen was determined by histological evaluation of hematoxylin–eosin stained slices. For some parameters, a high correlation was found: the correlation of T_1Gd vs Eq (r = 0.8095), T_1Gd vs I/Eq (r = −0.8441) and T_1Gd vs τ (r = 0.8469). The correlation of T₂ and ADC with selected biomechanical parameters was not statistically significant.

In conclusion, GAG content measured by dGEMRIC is highly related to the selected biomechanical properties of naturally degenerated articular cartilage. In contrast, T₂ and ADC were unable to estimate these properties. The results of the study imply that some MR parameters can non-invasively predict the biomechanical properties of degenerated articular cartilage.

Introduction

The articular cartilage can be thought of as a fiber-reinforced anisotropic composite material. The mechanical properties of articular cartilage are a function of the essential mechanical properties of the tissue components and the interplay of these components during loading. Articular cartilage is a bi-phasic material: the permeable solid section is represented by a solid matrix that consists of collagen fibers and proteoglycan molecules, and the fluid section is composed of extracellular water with dissolved ions and nutrients. Most of the fluid can move through the collagen network during loading. Experimental data from articular cartilage can be assessed by single-, bi- or multi-phasic models [1], [2], [3], [4].

The mechanical properties of articular cartilage arise from the complex structure and interactions of its biochemical constituents: mostly water, electrolytes, and a solid matrix composed primarily of collagen and proteoglycan. Laasanen et al. showed that collagen primarily controls the dynamic tissue response while proteoglycans affect more the static properties [5]. In the ex-vivo study published by Appleyard et al. there was found a strong and significant correlation between shear modulus and collagen content, whereas correlation between shear modulus and water content was not significant [6]. Stiffness and strength of articular cartilage tissue depended on the density and orientation of collagen fibers, and the type or amount of collagen cross-linking [7].

Various imaging methods have been applied to assess articular cartilage. These include standard radiography [8], CT arthrography [9], ultrasonography [10], and MR imaging [11], [12]. MRI offers excellent soft tissue contrast and multi-planar imaging capability, and therefore, it has become the method of choice to diagnose cartilage diseases. It also provides valuable information on composition and structural changes in cartilage. For determination of glycosaminoglycan (GAG) content, delayed gadolinium-enhanced MRI of cartilage (dGEMRIC) was developed and has been used to answer clinical questions [13], [14]. Since transversal relaxation time, T₂, is related to collagen matrix composition and is primarily influenced by collagen fiber orientation [15] and water content [16], T₂ has been successfully used for in vitro [17] and in vivo [18], [19] cartilage tissue assessment. Another useful parameter that reflects the functional properties of cartilage tissue is diffusivity. The idea of using gradients to make the MR imaging signal sensitive to the molecular motion of water was introduced by Stejskal and Tanner in 1965 [20], and currently, there are a few sequences based on diffusion weighting that have been successfully used for evaluation of cartilage function and its structure [21], [22], [23], [24].

The mechanical properties of articular cartilage can be evaluated on different levels, from an in vivo investigation of intact joints by indentation [25] to in vitro studies of explants [26] and to cellular and molecular levels [26]. For viscoelastic materials like cartilage, two biomechanical parameters play a role: firstly the instantaneous modulus (I), which describes the initial stiffness upon loading, and secondly the equilibrium modulus (Eq) which describes only the time independent share of the stiffness. The ratio between I and Eq gives a measure of the viscoelastic character. If I/Eq is close to unity the material is merely elastic, and is better termed spontaneous elastic. If I/Eq is greater than two, the material is strongly viscoelastic. For cartilage it may be said that the smaller I/Eq the more vital is the cartilage, which means it is able to better withstand static loading (carrying heavy loads for instance) as well as fast and extremely high load cycles (from jumping or running for instance). The mechanical properties of cartilage tissue are strongly related to the OA stage of cartilage [27]. Changes in the biomechanical properties of articular cartilage are one of the first signs of the tissue degeneration. For cartilage degeneration, it is necessary to know how cartilage maintains its functionality and how cartilage responds to the ever-changing mechanical environment. The biomechanical properties of articular cartilage can be assessed relatively precisely, however, ex-vivo [28] or invasively during an arthroscopic procedure [29]. MRI provides the possibility to non-invasively assess the biomechanical properties of articular cartilage provides MRI. In the past, several studies suggested the way how MRI parameters might be used for biomechanical parameters assessment [28], [30], [31]. However, to the best of our knowledge, the correlation between biomechanical properties and MR parameters of degenerated human articular cartilage stages has not been previously determined.

Therefore, in the presented study we hypothesized that biochemical parameters derived non-invasively from MRI as a non-invasive imaging method can predict the biomechanical properties of naturally degenerated human articular cartilage. Thus the aim of this work was to determine the correlation of MR relaxation times and diffusion constants at 3 T with an instantaneous and an equilibrium modulus.