Jukka S. Jurvelin

T2 relaxation reveals spatial collagen architecture in articular cartilage: A comparative quantitative MRI and polarized light microscopic study

It has been suggested that orientational changes in the collagen network of articular cartilage account for the depthwise T2 anisotropy of MRI through the magic angle effect. To investigate the relationship between laminar T2 appearance and collagen organization (anisotropy), bovine osteochondral plugs (N = 9) were T2 mapped at 9.4T with cartilage surface normal to the static magnetic field. Collagen fibril arrangement of the same samples was studied with polarized light microscopy, a quantitative technique for probing collagen organization by analyzing its ability to rotate plane polarized light, i.e., birefringence (BF). Depthwise variation of safranin O‐stained proteoglycans was monitored with digital densitometry. The spatially varying cartilage T2 followed the architectural arrangement of the collagen fibril network: a linear positive correlation between T2 and the reciprocal of BF was established in each sample, with r = 0.91 ± 0.02 (mean ± SEM, N = 9). The current results reveal the close connection between the laminar T2 structure and the collagen architecture in histologic zones. Magn Reson Med 46:487–493, 2001.

Comparison of the equilibrium response of articular cartilage in unconfined compression, confined compression and indentation

At mechanical equilibrium, articular cartilage is usually characterized as an isotropic elastic material with no interstitial fluid flow. In this study, the equilibrium properties (Youngs modulus, aggregate modulus and Poissons ratio) of bovine humeral, patellar and femoral cartilage specimens (n=26) were investigated using unconfined compression, confined compression, and indentation tests. Optical measurements of the Poissons ratio of cartilage were also carried out. Mean values of the Youngs modulus (assessed from the unconfined compression test) were 0.80+/-0.33, 0.57+/-0.17 and 0.31+/-0.18MPa and of the Poissons ratio (assessed from the optical test) 0.15+/-0.06, 0.16+/-0.05 and 0.21+/-0.05 for humeral, patellar, and femoral cartilages, respectively. The indentation tests showed 30-79% (p<0.01) higher Youngs modulus values than the unconfined compression tests. In indentation, values of the Youngs modulus were independent of the indenter diameter only in the humeral cartilage. The mean values of the Poissons ratio, obtained indirectly using the mathematical relation between the Youngs modulus and the aggregate modulus in isotropic material, were 0.16+/-0.06, 0.21+/-0.05, and 0.26+/-0.08 for humeral, patellar, and femoral cartilages, respectively. We conclude that the values of the elastic parameters of the cartilage are dependent on the measurement technique in use. Based on the similar values of Poissons ratios, as determined directly or indirectly, the equilibrium response of articular cartilage under unconfined and confined compression is satisfactorily described by the isotropic elastic model. However, values of the isotropic Youngs modulus obtained from the in situ indentation tests are higher than those obtained from the in vitro unconfined or confined compression tests and may depend on the indenter size in use.

Normal and pathological adaptations of articular cartilage to joint loading.

Joints are functional units that transmit mechanical loads between contacting bones during normal daily or specialized activities, e.g., sports. All components of the joint, i.e. articular cartilage, bone, muscles, ligaments/tendons and nerves, participate in load transmission. Failure in any of these components can cause joint malfunction, which, in turn, may lead to accumulation of damage in other joint components. Mechanical forces have great influence on the synthesis and rate of turnover of articular cartilage molecules, such as proteoglycans (PGs). Regular cyclic loading of the joint enhances PG synthesis and makes cartilage stiff. On the other hand, loading appears to have less evident effects on the articular cartilage collagen fibril network. Continuous compression of the cartilage diminishes PG synthesis and causes damage of the tissue through necrosis. The prevailing view is that osteoarthrosis (OA) starts from the cartilage surface through PG depletion and fibrillation of the superficial collagen network. It has also been suggested that the initial structural changes take place in the subchondral bone, especially when the joint is exposed to an impact type of loading. This in turn would create an altered stress pattern on joint surfaces, which leads to structural damage and mechanical failure of articular cartilage. The importance of the neuromuscular system to the initiation and progression of OA is still poorly understood. Many surgical extra‐ and intra‐articular procedures have been used for the treatment of OA. Although some of the new methods, such as autologous chondrocyte transplantation and mosaicplasty, have given good clinical results, it is reasonable to emphasize that the methods still are experimental and more controlled studies are needed.

Fibril reinforced poroelastic model predicts specifically mechanical behavior of normal, proteoglycan depleted and collagen degraded articular cartilage

Degradation of collagen network and proteoglycan (PG) macromolecules are signs of articular cartilage degeneration. These changes impair cartilage mechanical function. Effects of collagen degradation and PG depletion on the time-dependent mechanical behavior of cartilage are different. In this study, numerical analyses, which take the compression-tension nonlinearity of the tissue into account, were carried out using a fibril reinforced poroelastic finite element model. The study aimed at improving our understanding of the stress-relaxation behavior of normal and degenerated cartilage in unconfined compression. PG and collagen degradations were simulated by decreasing the Youngs modulus of the drained porous (nonfibrillar) matrix and the fibril network, respectively. Numerical analyses were compared to results from experimental tests with chondroitinase ABC (PG depletion) or collagenase (collagen degradation) digested samples. Fibril reinforced poroelastic model predicted the experimental behavior of cartilage after chondroitinase ABC digestion by a major decrease of the drained porous matrix modulus (-64+/-28%) and a minor decrease of the fibril network modulus (-11+/-9%). After collagenase digestion, in contrast, the numerical analyses predicted the experimental behavior of cartilage by a major decrease of the fibril network modulus (-69+/-5%) and a decrease of the drained porous matrix modulus (-44+/-18%). The reduction of the drained porous matrix modulus after collagenase digestion was consistent with the microscopically observed secondary PG loss from the tissue. The present results indicate that the fibril reinforced poroelastic model is able to predict specifically characteristic alterations in the stress-relaxation behavior of cartilage after enzymatic modifications of the tissue. We conclude that the compression-tension nonlinearity of the tissue is needed to capture realistically the mechanical behavior of normal and degenerated articular cartilage.

Microspectrophotometric quantitation of glycosaminoglycans in articular cartilage sections stained with Safranin O

SummaryA new microspectrophotometric method was developed for quantitation of glycosaminoglycans with Safranin O dye in articular cartilage matrix. From histological sections molar extinction coefficient of Safranin O was determined and used to measure the dye content of the sections. The amount of glycosaminoglycans was determined with depth of bovine articular cartilage by both gas chromatography and thin layer chromatography to calculate the fixed negative charge content. Comparison between the results revealed that binding of Safranin O to glycosaminoglycan polyanions was stoichiometric and showed minimal nonspecific staining. The method provides an accurate technique for quantitation and localization of fixed negative charge content of glycosaminoglycans in the articular cartilage matrix. Specific enzyme digestions enable detection of separate glycosaminoglycans.

Osteoporosis in adult patients with celiac disease

We investigated the bone mineral density (BMD) and prevalence of osteopenia and osteoporosis in adult celiac patients with varying disease states. In this cross-sectional study the data on the severity of celiac disease and BMD were collected from 77 celiac patients (28 newly diagnosed and 49 previously diagnosed celiac patients), and BMD results were compared with those of 157 control subjects matched for age, gender, and menopausal status. The celiac patients had significantly lower BMD than the control subjects at the lumbar spine (-6%) and femoral neck (-5%). The mean BMD did not differ significantly among celiac patients classified by severity of disease. Based on Z scores, 35% of the celiac patients and 17% of the control subjects had low BMDs for age at the lumbar spine (p = 0.005), whereas 31% of celiac patients and 16% of control subjects had Z scores of < or =-1 at the femoral neck (p = 0.01). Altogether, 26% of all celiac patients, but only 5% of control subjects, were classified as having osteoporosis (T score < or =-2.5 SD) at the lumbar spine (p = 0.03), whereas osteoporosis was rare at the femoral neck in both groups (3% vs. 1%, p = 1.00). Prevalence of osteopenia and osteoporosis was highest in newly diagnosed celiac patients and in patients with disease not in remission. A low 25-(OH)D vitamin concentration was a typical biochemical abnormality in our patients (64% of men and 71% of women). The main associated variables of low BMD were age (men), low serum vitamin D level, low body weight, and postmenopausal status (women). The present study suggests that celiac disease constitutes a risk factor for osteoporosis. This finding applies particularly to untreated and poorly treated patients.

Quantitative MR microscopy of enzymatically degraded articular cartilage.

Structural changes in bovine patellar articular cartilage, induced by component selective enzymatic treatments, were investigated by measuring tissue T2 relaxation at 9.4 T. This MRI parameter was compared with Youngs modulus, a measure of elastic stiffness and loadbearing ability of cartilage tissue. Collagenase was used to digest the collagen network and chondroitinase ABC to remove proteoglycans. Polarized light microscopy and digital densitometry were used to assess enzyme penetration after 44 hr of enzymatic digestion. T2 relaxation in superficial cartilage increased significantly only in samples treated with collagenase. A statistically significant decrease in Youngs modulus was observed in both enzymatically treated sample groups. These results confirm that T2 of articular cartilage is sensitive to the integrity of collagen in the extracellular matrix. Nonetheless, it does not appear to be an unambiguous indicator of cartilage stiffness, which is significantly impaired in osteoarthrosis. Magn Reson Med 43:676–681, 2000.

Importance of the superficial tissue layer for the indentation stiffness of articular cartilage

Indentation testing is a widely used technique for nondestructive mechanical analysis of articular cartilage. Although cartilage shows an inhomogeneous, layered structure with anisotropic mechanical properties, most theoretical indentation models assume material homogeneity and isotropy. In the present study, quantitative polarized light microscopy (PLM) measurements from canine cartilage were utilized to characterize thickness and structure of the superficial, collageneous tissue layer as well as to reveal its relation to experimental indentation measurements. In addition to experimental analyses, a layered, transversely isotropic finite element (FE) model was developed and the effect of superficial (tangential) tissue layer with high elastic modulus in the direction parallel to articular surface on the indentation response was studied. The experimental indentation stiffness was positively correlated with the relative thickness of the superficial cartilage layer. Also the optical retardation, which reflects the degree of parallel organization of collagen fibrils as well as collagen content, was related to indentation stiffness. FE results indicated effective stiffening of articular cartilage under indentation due to high transverse modulus of the superficial layer. The present results suggest that indentation testing is an efficient technique for the characterization of the superficial degeneration of articular cartilage.

Spatial assessment of articular cartilage proteoglycans with Gd-DTPA-enhanced T1 imaging

In Gd‐DTPA‐enhanced T1 imaging of articular cartilage, the MRI contrast agent with two negative charges is understood to accumulate in tissue inversely to the negative charge of cartilage glycosaminoglycans (GAGs) of proteoglycans (PGs), and this leads to a decrease in the T1 relaxation time of tissue relative to the charge in tissue. By assuming a constant relaxivity for Gd‐DTPA in cartilage, it has further been hypothesized that the contrast agent concentration in tissue could be estimated from consecutive T1 measurements in the absence or presence of the contrast agent. The spatial sensitivity of the technique was examined at 9.4 T in normal and PG‐depleted bovine patellar cartilage samples. As a reference, spatial PG concentration was assessed with digital densitometry from safranin O‐stained cartilage sections. An excellent linear correlation between spatial optical density (OD) of stained GAGs and T1 with Gd‐DTPA was observed in the control and chondroitinase ABC‐treated cartilage specimens, and the MR parameter accounted for approximately 80% of the variations in GAG concentration within samples. Further, the MR‐resolved Gd‐DTPA concentration proved to be an even better estimate for PGs, with an improved correlation. However, the linear relation between MR parameters and PG concentration did not apply in the deep tissue, where MR measurements overestimated the PG content. While the absolute [Gd‐DTPA] determination may be prone to error due to uncertainty of relaxivity in cartilage, or to other contributing factors such as variations in tissue permeability, the experimental evidence highlights the sensitivity of this technique to reflect spatial changes in cartilage PG concentration in normal and degenerated tissue. Magn Reson Med 48:640–648, 2002.

Monitoring of Periprosthetic BMD After Uncemented Total Hip Arthroplasty with Dual-Energy X-Ray Absorptiometry—a 3-Year Follow-Up Study

Insertion of a metallic implant into the femur changes bone loading conditions and results in remodeling of femoral bone. To quantify changes in bone mass after uncemented total hip arthroplasty (THA), we monitored femoral bone with dual‐energy X‐ray absorptiometry (DXA). The periprosthetic bone mineral density (BMD) was measured with Lunar DPX densitometry in seven Gruen zones and the total periprosthetic area at scheduled time intervals in 22 patients during a 3‐year follow‐up. BMD decreased significantly almost in all Gruen zones during the first 3 months, ranging from 3.4% to 14.4% (p < 0.05 to p < 0.001). At the end of the first year, the most remarkable decrease in BMD was found in the calcar (zone 7; −22.9%). During the second postoperative year, a slight restoration of periprosthetic bone mass was recorded. During the third year, no significant changes in BMD were found. The preoperative BMD was the only factor that was significantly related to the periprosthetic bone loss. Clearly, the early periprosthetic bone loss noticed during the 3 months after THA is caused by mainly limited weight bearing to the operated hip and stress shielding. We suggest that the restoration of bone mass is a sign of successful osteointegration between bone and metallic implant. DXA is a suitable tool to follow the bone response to prosthetization and will increase our knowledge on the behavior of bone after THA.