Real-time ultrasound analysis of articular cartilage degradation in vitro

Ultrasound in Med. & Biol., Vol. 28, No. 4, pp. 519 –525, 2002 Copyright © 2002 World Federation for Ultrasound in Medicine & Biology Printed in the USA. All rights reserved 0301-5629/02/$–see front matter

PII: S0301-5629(02)00480-5

● Original Contribution REAL-TIME ULTRASOUND ANALYSIS OF ARTICULAR CARTILAGE DEGRADATION IN VITRO HEIKKI J. NIEMINEN*†, JUHA TO¨ YRA¨ S*‡, JARNO RIEPPO†, MIIKA T. NIEMINEN†, JANI HIRVONEN*, RAMI KORHONEN*† and JUKKA S. JURVELIN*‡ *Department of Clinical Physiology and Nuclear Medicine, Kuopio University Hospital and University of Kuopio, Kuopio, Finland; Departments of †Anatomy and ‡Applied Physics, University of Kuopio, Kuopio, Finland (Received 25 January 2001; in final form 17 December 2001)

Abstract—The sensitivity of the reflection coefficient, attenuation and velocity to the enzymatic degradation of bovine patellar cartilage was evaluated in real-time with high-frequency ultrasound (US) (29.4 MHz). These parameters were estimated from the radiofrequency (RF) signal, which was recorded at 5-min intervals during the digestion of the tissue by collagenase or by trypsin. The coefficient of reflection at cartilage surface decreased by 78.5% and 10.5% (p < 0.05) after 6 h of exposure to collagenase and 4 h of exposure to trypsin, respectively. During the trypsin digestion, the attenuation in cartilage increased by 0.274 dB/mm (p < 0.05) and the velocity decreased by 7 m/s (p < 0.05). The coefficient of reflection at the cartilage surface was the most sensitive acoustic parameter to the enzymatic degradation of cartilage and may be the easiest to implement for clinical diagnosis of cartilage quality. US velocity was found to be insensitive to degradation. The small difference in mean velocity between the control and degraded cartilage suggests that a constant predefined US velocity value can be used to obtain diagnostically acceptable measurement of the cartilage thickness. (E-mail: [email protected]) © 2002 World Federation for Ultrasound in Medicine & Biology. Key Words: Ultrasound, Articular cartilage, Enzymatic digestion, Osteoarthrosis.

invasive imaging techniques such as X-ray and magnetic resonance imaging (MRI), and invasive techniques such as arthroscopy and mechanical indentation (Lyyra et al. 1995). However, X-ray techniques are insensitive for evaluation of pathologic changes because they cannot reveal structure or properties of soft tissues. The resolution of clinical MRI techniques, on the other hand, is insufficient to reveal superficial tissue damage. Conventional arthroscopy relies on the qualitative, visual evaluation and reveals only macroscopic tissue changes of progressed disease. The mechanical indentation technique lacks information on the cartilage thickness, making determination of intrinsic tissue stiffness impossible (Hayes et al. 1972; Mak et al. 1987). High-frequency US may provide sensitive means for a quantitative evaluation of the structural and functional properties of articular cartilage (Adams and Wallace 1991; Agemura et al. 1990; Cherin et al. 1998; Kim et al. 1995; Lefebvre et al. 1998; Saied et al. 1997; Senzig et al. 1992; To¨yra¨s et al. 1999; Youn et al. 1999). Morphologic changes in cartilage and subchondral bone induced by experimental OA were detected using a pulse-echo technique at 50 MHz (Cherin et al. 1998;

INTRODUCTION Articular cartilage is a connective tissue that protects articulating bones against friction and contact loads. It is generally described as a biphasic tissue constituted of a solid matrix and of an interstitial fluid representing 70% to 80% of the total volume (Muir 1980). The matrix consists of a fibrous network mainly of type II collagen, which traps electrolytes and macromolecules, proteoglycans (PGs). These hydrophilic macromolecules bind large amounts of water in the tissue. Collagens and PGs are the most important structural components of the cartilage, and the interaction between these constituents is essential for the mechanical integrity of the tissue (Broom and Poole 1983). Osteoarthrosis (OA) is a severe musculoskeletal disease, commonly diagnosed among elderly people. For successful treatment of the disease, early detection of the pathologic changes in cartilage is essential. Current diagnostic modalities of cartilage degradation include nonAddress correspondence to: Juha To¨yra¨s, Department of Applied Physics, University of Kuopio, POB 1627, 70211 Kuopio, Finland. E-mail: [email protected] 519

520

Ultrasound in Medicine and Biology

Saied et al. 1997). In a recent study, ultrasonic B-mode imaging was found to be a promising tool for detecting and grading focal cartilage lesions (Disler et al. 2000). The suitability of US for the measurement of cartilage thickness has also been studied (Jurvelin et al. 1995; Modest et al. 1989; Myers et al. 1994; No¨ tzli et al. 1994). To be able to measure tissue thickness reliably, accurate information of US velocity in the tissue is required. In addition to morphologic parameters, such as thickness, US enables the registration of acoustic parameters, including the reflection, backscatter and attenuation coefficients, which may serve in the evaluation of functional properties of the tissue (Cherin et al. 1998; To¨ yra¨ s et al. 1999). The present study focused on the sensitivity of ultrasonic parameters to reveal the foremost structural changes in the superficial layer of cartilage. The reflection coefficient at the cartilage surface, the attenuation and the velocity of sound were quantitated in real-time using high-frequency US during the enzymatic degradation of bovine patellar cartilage. Specifically, collagenase and trypsin were used to digest collagen and PGs, respectively (Harris et al. 1972; Shingleton et al. 1996), to reveal interrelationships between cartilage composition and acoustic parameters. The current experimental setup enabled accurate ultrasonic real-time measurements to assess changes in the relative and absolute acoustic parameters during digestion. MATERIALS AND METHODS Cartilage samples Knee joints of 1- to 3-year-old bulls (n ⫽ 18) were obtained from the local slaughterhouse (Atria Oyj, Kuopio, Finland). One sample from each patella was prepared under sterile conditions. The extracted patella was fixed with a clamp and an osteochondral plug from the lateral upper quadrant was drilled using a hollow bit (diameter ⫽ 16 mm), and subsequently removed using an autopsy saw (Stryker Autopsy Saw 868, Stryker Europe bv, Uden, Netherlands). The bone was trimmed with a diamond saw (Buehler Isomet Low Speed Saw, Buehler Ltd, Lake Bluff, IL), leaving about 1 mm of subchondral bone beneath the cartilage. The final osteochondral sample (diameter ⫽ 9 mm) was subsequently prepared with an autopsy punch. During preparation, the sample was kept moist with Gibco BRL Dulbecco’s phosphate-buffered saline (PBS) (Life Technologies, Paisley, Scotland, UK) (To¨ yra¨ s et al. 1999). The samples were divided into three groups: a control group (n ⫽ 6), collagenase-digested group (n ⫽ 6) and a trypsin-digested group (n ⫽ 6). The control samples were incubated without enzyme for 6 h at 37°C, 5% CO2 atmosphere in Gibco BRL MEM (Life Technolo-

Volume 28, Number 4, 2002

gies) with 70 ␮g/ml L-ascorbic acid (Sigma Chemical Co., St. Louis, MO), 100 ␮g/mL streptomycin (Sigma), 100 U/ml penicillin (PAA Laboratories, Linz, Austria), 200 mM L-glutamine (PAA Laboratories) and 2 ␮g/ml fungiside (Gibco BRL Fungizone, Life Technologies). For enzymatic digestion, the medium was supplemented with 30 U/ml collagenase type VII (C 0773, Sigma) for type II collagen degradation, and 200 ␮g/mL trypsin (T 0646, Sigma) for PG digestion. Trypsin has also been reported to cleave minor amounts of collagen (Harris et al. 1972). The samples and solutions were separately warmed up to 37°C in the incubator before treatments. After US assessment, the samples were rinsed with PBS and immersed in PBS with enzyme inhibitors: 5 mM ethylenediamine tetraacetic acid, EDTA (Riedel-deHaen, Seelze, Germany); 5 mM benzamide-HCl (Sigma); and 5 ␮g/l phenylmethylsulfonyl fluoride (Sigma). The enzymatical degradations were conducted to induce degradation of superficial collagens or PGs, which is typical of early OA (Buckwalter and Mankin 1997). Ultrasound measurements An UltraPAC System (Physical Acoustic Corporation, Princeton, NJ) equipped with a focused broadband 29.4-MHz (18.8 to 40.0 MHz, ⫺6 dB) pulse-echo transducer (diameter ⫽ 6.35 mm, radius of curvature ⫽ 25.40 mm, model PZ25-0.25⬙-SU-R1.00⬙, Panametrics, Waltham, MA) connected to a PAC IPR-100 Pulser receiver board (Physical Acoustic Corporation) was used for measurements. The system was controlled by custom software developed with LabVIEW (version 5.1.1, National Instruments, Austin, TX). The transducer was attached to a stand and a sample holder containing the cell culture medium was placed on top of it (Fig. 1). The subchondral bone was fixed to the lid of the holder with cyanoacrylate glue, so that the cartilage surface was perpendicular to the US beam and facing the transducer. This geometry enabled the determination of acoustic properties and the detection of the changes in cartilage thickness during digestion. When the specimen, transducer, sample holder and solution had reached a constant temperature of 37°C in the incubator, the plug was positioned in the holder so that the echo reflected at the cartilage surface was maximal. For each sample, the US signal reflected from the tissue was acquired at 5-min intervals through the incubation period (6 or 4 h), digitized at 500 MHz and recorded. At each time, 20 acquisitions were performed and the signals were averaged. Bandpass filtering (6 to 55 MHz) was used to remove noise from the signal. After incubation, the sample was removed from the incubator, rinsed and immersed in buffer with enzyme inhibitors. Following this, thickness measurements and histologic analyses were conducted.

Ultrasonic evaluation of articular cartilage ● H. J. NIEMINEN et al.

521

Fig. 2. Schematic representation of the cartilage structure and of the US transmission/reflection at the different interfaces between coupling medium, cartilage and bone (left). Typical US RF signal registered from the cartilage (right).

total cartilage thickness and the depth of penetration of the trypsin in the sample were measured from DD images. Fig. 1. Schematic presentation of the setup for the real-time US pulse-echo measurements of articular cartilage. A sample holder was attached on top of the transducer. The suchondral plug was glued on a lid that was positioned on top the holder for 6 or 4 h incubation. The cartilage surface was positioned perpendicular to US beam at a constant distance from the transducer. Cell culture medium, added into the holder, was used for immersion of the sample and as a coupling medium for propagation of US. The measurement took place in an incubator.

Thickness measurements and histologic analysis Cartilage thickness was measured at the location of the US acquisition, using the needle-probe technique (Jurvelin et al. 1995). In this technique, a sharp needle is pushed through the cartilage with a constant speed by a computer-controlled actuator, and the force by which the tissue resists penetration is recorded as a function of the penetration distance. The distance between cartilage surface and cartilage-bone-interface (i.e., tissue thickness) was determined from the variations of the mechanical resistance. The birefringence property of the collagen network was exploited to assess qualitatively the fibril organization and the degree of collagen degeneration using polarized light microscopy (PLM) (Arokoski et al. 1996; Kiraly et al. 1997; Modis 1991). The distribution of PGs in the cartilage was evaluated using digital densitometry (DD) (Panula et al. 1998). The cartilage sections were stained with safranin O, which binds stoichiometrically to glycosaminoglycan polyanions of PGs (Kiviranta et al. 1985), and the local PG concentration was determined quantitatively by measurement of the optical density of histologic sections (Kiraly et al. 1996a, 1996b). Also, the

Data analysis The mean US velocity in cartilage was calculated from the thickness measurement, which was obtained using the needle-probe technique, and the time of flight (TOF) between the echoes reflected from the mediumcartilage interface and cartilage-bone interface, respectively. The TOF was measured from the RF signal acquired at 5-min intervals during the incubation period. The cartilage surface was detected from the first zero crossing-point that preceded the first data sample of reflection amplitude higher than the level of noise. Similarly, the cartilage-bone interface was detected from the first zero crossing-point that preceded the first data sample of amplitude higher than the level of the signal from the internal cartilage (Fig. 2). Consequently, US velocity v (m/s) was calculated using the equation: v⫽

x NP , t US

(1)

where xNP is the cartilage thickness measured with the needle-probe and tUS ⫽ TOF/2. The coefficient of reflection R1 at the interface between the cell culture medium and the cartilage was calculated as: R1 ⫽

A1 , A0

(2)

where A1 is the peak-to-peak amplitude of the echo reflected from the medium-cartilage interface. A0 is the peak-to-peak amplitude of the pulse reflected from a

522

Ultrasound in Medicine and Biology

Volume 28, Number 4, 2002

Table 1. Reflection coefficient R1, attenuation ␣ (dB/mm), and US velocity v (m/s) at the beginning and after enzymatic digestion with collagenase (6 h) or trypsin (4 h) (mean ⫾ SD) ␣ (dB/mm)

R1 0h

6 h or 4 h

⌬(%)

0h

6 h or 4 h

Control (6 h) 0.0280 ⫾ 0.0022 0.0277 ⫾ 0.0024 ⫺1.0 9.791 ⫾ 1.065 9.760 ⫾ 1.083 Collagenase (6 h) 0.0243 ⫾ 0.0076 0.0052 ⫾ 0.0021 ⫺78.5* 11.494 ⫾ 2.904 11.413 ⫾ 2.910 Trypsin (4 h) 0.0214 ⫾ 0.0082 0.0191 ⫾ 0.0087 ⫺10.5* 11.152 ⫾ 1.327 11.426 ⫾ 1.278

v (m/s) ⌬(dB/mm) ⫺0.031 ⫺0.081 ⫹0.274*

0h

6 h or 4 h

⌬(%)

1634 ⫾ 101 1637 ⫾ 101 ⫹0.2* 1666 ⫾ 131 1662 ⫾ 129 ⫺0.2 1622 ⫾ 194 1615 ⫾ 195 ⫺0.4*

* Significantly different ( p ⬍ 0.05) from the value at t ⫽ 0 h (Wilcoxon signed ranks test for paired samples).

perfect reflector located at the same distance from the transducer as the surface of the cartilage. A medium-air interface was used as perfect reflector to determine A0. The attenuation coefficient ␣ (dB/mm) was calculated as follows (Brown 1997):

␣⫽

冉

冊

R 2A 1 4.343 共1 ⫺ R 12兲 , ln h R 1A 2

(3)

where h is the cartilage thickness, A1 and A2 are the peak-to-peak amplitudes of the echoes reflected from the medium-cartilage interface and cartilage-bone interface, respectively. R1 is the reflection coefficient defined in eqn (2) and R2 is the reflection coefficient at the cartilage-bone interface, which was assumed to be the same for all samples (R2 ⫽ 0.374) (To¨ yra¨ s et al. 1999; Wells 1977). The data were processed and acoustic parameters were calculated using a custom application developed with Matlab (version 5.3.1, Mathworks Inc., Natick, MA). The M-mode image of trypsin-digested samples was created by plotting the envelope of US A-mode signal, calculated as the absolute value of the Hilbert transform, as a function of incubation time. Due to a limited amount of samples, the nonparametric Wilcoxon signed ranks test for paired samples (SPSS 8.0.1, SPSS Inc., Chigaco, IL) was used to estimate the significance of the changes in the acoustic parameters during the experiment. Mann–Whitney U test (SPSS) was used to assess the difference in US parameters between digested and control groups at the beginning of incubation.

coefficient of medium-cartilage interface of trypsin-digested samples decreased slightly, but significantly (⫺10.5%, p ⬍ 0.05, Table 1, Fig. 3). Attenuation In the control and collagenase groups, the decrease in attenuation coefficient ␣ during the incubation was 0.031 dB/mm and 0.081 dB/mm, respectively, but was not statistically significant (Table 1, Fig. 3). During trypsin digestion, a significant increase in the attenuation was observed, 0.274 dB/mm (p ⬍ 0.05, Table 1, Fig. 3). The attenuation values for collagenase- and trypsin-digested samples at t ⫽ 0 h were 1.703 dB/mm and 1.361 dB/mm, respectively, greater than the equivalent control value 9.791 ⫾ 1.065 dB/mm; however, these differences were statistically insignificant.

RESULTS Reflection coefficient In the control group, the change in the reflection coefficient during incubation was statistically insignificant (⫺1.0%, Table 1, Fig. 3). During 6-h collagenase digestion, US reflection coefficient from the cell culture medium-cartilage interface decreased significantly (⫺78.5%, p ⬍ 0.05, Table 1, Fig. 3). The reflection

Fig. 3. (a) Mean reflection coefficient at the cartilage surface, (b) mean attenuation coefficient, and (c) mean velocity during the collagenase or trypsin digestions. Statistically significant changes in the reflection coefficient were observed in collagenase- and trypsin-digested cartilage. Attenuation change was found to be statistically significant only in the trypsin-digested group. US velocity was found to change significantly only in the trypsin-digested and control groups.

Ultrasonic evaluation of articular cartilage ● H. J. NIEMINEN et al.

Fig. 4. M-mode visualization of the trypsin penetration in the cartilage as a function of digestion time (top). Comparison with a light microscopy image of the safranin O-stained cartilage sample (bottom). The trypsin penetration front in the M-mode image is indicated with arrows.

Velocity US velocities in control and trypsin-digested samples experienced small, but statistically significant changes of ⫹ 3 m/s (⫹ 0.2%, p ⬍ 0.05, Table 1, Fig. 3) and ⫺7 m/s (⫺0.4%, p ⬍ 0.05, Table 1, Fig. 3), respectively. In the collagenase-digested group, velocity decreased with 4 m/s (⫺0.2%, Table 1, Fig. 3), but the change was statistically insignificant. US velocity in collagenase- and trypsin-digested samples at t ⫽ 0 h differed with ⫹ 32 m/s and ⫺12 m/s, respectively, from the equivalent control value 1634 ⫾ 101 m/s, but this was not statistically significant. Progression of the trypsin front was sensitively detected with US (Fig. 4). Echo amplitude from the penetration front and the maximum gradient of PG concentration, as determined by DD, matched spatially and correlated significantly (r ⫽ 0.917, n ⫽ 6, p ⬍ 0.05). The effect of degradation on cartilage thickness was found to be negligible in all groups. Histology findings As seen in PLM images (Fig. 5), no changes were observed in the collagen network of control or trypsindigested samples, and only minor changes in birefringence were observed in the collagenase-digested samples. Severe PG depletion in the superficial and intermediate zones was observed after trypsin digestion (Fig. 5), but no PG loss was observed in control or collagenasedigested specimens. The mean cartilage thickness for trypsin-digested

523

Fig. 5. Polarized light microscopic (PLM) (left) and light microscopy images of safranin O-stained cartilage samples (right). (a) Control, (b) collagenase-digested and (c) trypsindigested samples. No changes were observed in the collagen network of control or trypsin-digested samples, and only minor changes in superficial cartilage birefringence were observed in the collagenase-digested samples. Severe PG depletion was observed after trypsin digestion, but no depletion was observed in the control or collagenase-digested specimens.

samples was 1.63 ⫾ 0.15 mm and the depth of the trypsin penetration after 4-h incubation was 0.97 ⫾ 0.05 mm, as measured from the histologic sections. The cartilage thicknesses (mean ⫾ SD) in each group, as measured with the needle-probe technique, were 1.95 ⫾ 0.48 mm for the control group, 1.68 ⫾ 0.23 mm for the collagenase-digested group and 1.71 ⫾ 0.33 mm for the trypsindigested group. DISCUSSION AND SUMMARY High-frequency US was used to evaluate the acoustic properties of articular cartilage in vitro, and to investigate the sensitivity of quantitative US parameters to cartilage degeneration. For the first time, the variations of the acoustic properties of articular cartilage were evaluated in real-time during the degradation of the tissue. Enzymatic degradations, used in this study, induced minor changes on cartilage structure and composition (Lyyra et al. 1999). These changes were specific compared to other common OA models (LeRoux et al. 2000; Setton et al. 1993). Thus, when studying cartilage structure-function relations, the enzymatic model appears as ideal. In contrast to in vivo models, the response of treatments can be easily adjusted by selection of enzymes and duration of treatments. Therefore, our experimental model induced superficial degeneration of the cartilage, which is, according to Buckwalter and Mankin (1997), typical of early OA. The reflection coefficient was found to be very sensitive to immediate changes induced by the collage-

524

Ultrasound in Medicine and Biology

nase in the superficial layer of the cartilage. The decrease of the reflection coefficient induced by trypsin was significantly smaller than that induced by collagenase. During PG digestion by trypsin, minor cleavage of collagen takes place (Harris et al. 1972). Secondarily, as qualitatively described (Pellaumail et al. 1998), but not quantitatively confirmed (Lyyra et al. 1999), changes in the density and size of collagen fibers may occur. However, even if present, these changes are minor without strong effects on acoustic reflection. The reflection coefficient is related to the difference in acoustic impedances between cartilage and coupling medium, and is also sensitive to the roughness of the interface between these two media (Adler et al. 1992; Chiang et al. 1994, 1997). Because neither collagenase-digested nor trypsin-digested samples showed any macroscopic fibrillation of the surface, the decrease of the reflection coefficient could primarily be associated with the variation of acoustic impedance of the cartilage. The reflection coefficient was revealed to be a very sensitive probe that can be used to assess the integrity of collagen network in the superficial cartilage layer. This finding is important because the injury of the collagen network is hypothesized to represent the “point of no return” in the process of OA progression (Buckwalter and Mankin 1997). In the beginning of the treatment, slight but insignificant differences in the cartilage acoustic properties between experimental groups were detected. In this study, mean US velocities from 1622 m/s to 1666 m/s in nondigested cartilage were measured. These values agree with the earlier mean values measured for bovine or human cartilage, 1654 to 1765 m/s (Agemura et al. 1990; Kolmonen et al. 1995; Myers et al. 1995; To¨ yra¨ s et al. 1999; Youn et al. 1999). It has been suggested that US velocity is slightly greater in normal (1658 m/s) than in OA cartilage (1581 m/s) (Myers et al. 1995), and that digestion of PGs reduces the US velocity in cartilage (Youn et al. 1999). In this study, only trypsin-digested cartilage showed a statistically significant reduction in US velocity (at t ⫽ 4 h) compared to that in the beginning of incubation (t ⫽ 0 h). However, the small difference in mean velocity through the total thickness of the cartilage between control and degraded cartilage suggests that a constant predefined US velocity value can be used to obtain diagnostically acceptable measurement of the cartilage thickness. In the current work, mean attenuation values for normal cartilage were between 9.791 and 11.494 dB/mm which, along with the earlier values of 2.780 to 6.515 dB/mm at 10 to 40 MHz (Senzig et al. 1992) and 92 to 147dB/mm at 100 MHz (Agemura et al. 1990), suggest a significant frequency-dependence of the attenuation. Attenuation, as determined in this study, is appropriate only for determining changes because the absolute value of

Volume 28, Number 4, 2002

attenuation can be controlled with the selection of the reflection coefficient value at the cartilage-bone interface. Contrary to an earlier study (To¨ yra¨ s et al. 1999), a statistically significant increase in US attenuation after PG degradation was observed. Several reasons can be suggested for this disagreement. First, in our previous study, PG depletion was induced by chondroitinase ABC which caused more superficial depletion than trypsin. Second, because all US measurements for each sample were performed at the same site and the sample remained untouched at all stages of the enzymatic treatment, the measurement geometry utilized in the current study enabled a more sensitive detection of the changes in acoustic properties than earlier. Third, the increase in attenuation, as measured in the present study, may also be due to an extra acoustical interface induced by trypsin. Recently, an US indentation technique (Zheng and Mak 1996) was used to characterize changes in the acoustic and mechanical properties of articular cartilage after enzymatic cleavage of PGs with trypsin (Youn et al. 1999). Their study reported a higher decrease in US velocity during trypsin treatment than revealed in the present study. In conclusion, high-frequency US is able to provide valuable quantitative information on the variations in composition and structure of the cartilage, and may be used in combination with arthroscopic indentation as a complementary tool for the evaluation of the structural and functional integrity of the tissue. Acknowledgements—Financial support from the Technology Development Center (TEKES), Finland, Kuopio University Hospital, Kuopio, Finland (EVO 5103), the Graduate School for Musculoskeletal Diseases, Finland and European Union (BMH4-CT97 to 2437) is acknowledged. Atria Oyj, Kuopio, Finland is acknowledged for material support. Finally, the authors thank Professor Heikki J. Helminen, University of Kuopio, Finland, for constructive criticism.

REFERENCES Adams ME, Wallace CJ. Quantitative imaging of osteoarthritis. Semin Arthritis Rheum 1991;20:26 –39. Adler RS, Dedrick DK, Laing TJ, et al. Quantitative assessment of cartilage surface roughness in osteoarthritis using high frequency ultrasound. Ultrasound Med Biol 1992;18:51–58. Agemura DH, O’Brien WD Jr, Olerud JE, Chun LE, Eyre DE. Ultrasonic propagation properties of articular cartilage at 100 MHz. J Acoust Soc Am 1990;87:1786 –1791. Arokoski JP, Hyttinen MM, Lapvetelainen T, 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. Ann Rheum Dis 1996;55:253–264. Broom ND, Poole CA. Articular cartilage collagen and proteoglycans. Their functional interdependency. Arthritis Rheum 1983;26:1111– 1119. Brown AE. Rationale and summary of methods for determining ultrasonic properties of materials at Lawrence Livermore National Laboratory. National Technical Information Service, U.S. Department of Commerce, Springfield: 1997. Buckwalter JA, Mankin HJ. Articular cartilage, Part II. Degeneration

Ultrasonic evaluation of articular cartilage ● H. J. NIEMINEN et al. and osteoarthritis, repair, regeneration, and transplantation. J Bone Joint Surg Am 1997;79:612– 632. Cherin E, Saied A, Laugier P, Netter P, Berger G. Evaluation of acoustical parameter sensitivity to age-related and osteoarthritic changes in articular cartilage using 50-MHz ultrasound. Ultrasound Med Biol 1998;24:341–354. Chiang EH, Adler RS, Meyer CR, et al. Quantitative assessment of surface roughness using backscattered ultrasound: The effects of finite surface curvature. Ultrasound Med Biol 1994;20:123–135. Chiang EH, Laing TJ, Meyer CR, et al. Ultrasonic characterization of in vitro osteoarthritic articular cartilage with validation by confocal microscopy. Ultrasound Med Biol 1997;23:205–213. Disler DG, Raymond E, May DA, Wayne JS, McCauley TR. Articular cartilage defects: In vitro evaluation of accuracy and interobserver reliability for detection and grading with US. Radiology 2000;215: 846 – 851. Harris ED Jr, Parker HG, Radin EL, Krane SM. Effects of proteolytic enzymes on structural and mechanical properties of cartilage. Arthritis Rheum 1972;15:497–503. Hayes WC, Keer LM, Herrmann G, Mockros LF. A mathematical analysis for indentation tests of articular cartilage. J Biomech 1972;5:541–551. Jurvelin JS, Rasanen T, Kolmonen P, Lyyra T. Comparison of optical, needle probe and ultrasonic techniques for the measurement of articular cartilage thickness. J Biomech 1995;28:231–235. Kim HK, Babyn PS, Harasiewicz KA, et al. Imaging of immature articular cartilage using ultrasound backscatter microscopy at 50 MHz. J Orthop Res 1995;13:963–970. Kiraly K, Hyttinen MM, Lapvetelainen T, et al. Specimen preparation and quantification of collagen birefringence in unstained sections of articular cartilage using image analysis and polarizing light microscopy. Histochem J 1997;29:317–327. Kiraly K, Lammi M, Arokoski J, et al. Safranin O reduces loss of glycosaminoglycans from bovine articular cartilage during histological specimen preparation. Histochem J 1996a;28:99 –107. Kiraly K, Lapvetelainen T, Arokoski J, et al. Application of selected cationic dyes for the semiquantitative estimation of glycosaminoglycans in histological sections of articular cartilage by microspectrophotometry. Histochem J 1996b;28:577–590. Kiviranta I, Jurvelin J, Tammi M, Saamanen AM, Helminen HJ. Microspectrophotometric quantitation of glycosaminoglycans in articular cartilage sections stained with safranin O. Histochemistry 1985;82:249 –255. Kolmonen P, Lyyra T, Jurvelin JS. Experimental comparison of acoustic and mechanical properties of bovine knee articular cartilage. Trans Orthop Res Soc 1995;20:513. Lefebvre F, Graillat N, Cherin E, Berger G, Saied A. Automatic three-dimensional reconstruction and characterization of articular cartilage from high-resolution ultrasound acquisitions. Ultrasound Med Biol 1998;24:1369 –1381. LeRoux MA, Arokoski J, Vail TP, et al. Simultaneous changes in the mechanical properties, quantitative collagen organization, and proteoglycan concentration of articular cartilage following canine meniscectomy. J Orthop Res 2000;18:383–392. Lyyra T, Arokoski JP, Oksala N, et al. Experimental validation of arthroscopic cartilage stiffness measurement using enzymatically degraded cartilage samples. Phys Med Biol 1999;44:525–535.

525

Lyyra T, Jurvelin J, Pitkanen P, Vaatainen U, Kiviranta I. Indentation instrument for the measurement of cartilage stiffness under arthroscopic control. Med Eng Phys 1995;17:395–399. Mak AF, Lai WM, Mow VC. Biphasic indentation of articular cartilage. I. Theoretical analysis. J Biomech 1987;20:703–714. Modest VE, Murphy MC, Mann RW. Optical verification of a technique for in situ ultrasonic measurement of articular cartilage thickness. J Biomech 1989;22:171–176. Modis L. Organization of the extracellular matrix: A polarization microscopic approach. Boca Raton, FL: CRC Press, 1991. Muir H. The chemistry of the ground substance of joint cartilage. In: Sokoloff L, ed. The joints and synovial fluid. New York: Academic Press, 1980:27–94. Myers S, Dines K, Albrecht M, et al. Assessment by high frequency ultrasound (HFU) of the thickness and subsurface characteristics of normal and osteoarthritic human cartilage. Trans Orthop Res Soc 1994;19:215. Myers SL, Dines K, Brandt DA, Brandt KD, Albrecht ME. Experimental assessment by high frequency ultrasound of articular cartilage thickness and osteoarthritic changes. J Rheumatol 1995;22: 109 –116. No¨ tzli H, Wilson A, Forster F, Clark J. Automated ultrasound measurements can map contour and thickness of articular cartilage. Trans Orthop Res Soc 1994;19:409. Panula HE, Hyttinen MM, Arokoski JP, et al. Articular cartilage superficial zone collagen birefringence reduced and cartilage thickness increased before surface fibrillation in experimental osteoarthritis. Ann Rheum Dis 1998;57:237–245. Pellaumail B, Loeuille D, Watrin A, et al. Correlation of high frequency ultrasound backscatter with cartilage matrix constituents. IEEE Ultrason Sympos 1998;2:1463–1466. Saied A, Cherin E, Gaucher H, et al. Assessment of articular cartilage and subchondral bone: Subtle and progressive changes in experimental osteoarthritis using 50 MHz echography in vitro. J Bone Miner Res 1997;12:1378 –1386. Senzig DA, Forster FK, Olerud JE. Ultrasonic attenuation in articular cartilage. J Acoust Soc Am 1992;92:676 – 681. Setton LA, Mow VC, Muller FJ, Pita JC, Howell DS. Altered structurefunction relationships for articular cartilage in human osteoarthritis and an experimental canine model. Agents Actions 1993; 39(Suppl.):27– 48. Shingleton WD, Hodges DJ, Brick P, Cawston TE. Collagenase: A key enzyme in collagen turnover. Biochem Cell Biol 1996;74:759 –775. To¨ yra¨ s J, Rieppo J, Nieminen MT, Helminen HJ, Jurvelin JS. Characterization of enzymatically induced degradation of articular cartilage using high frequency ultrasound. Phys Med Biol 1999;44: 2723–2733. Wells PNT. Biomedical ultrasonics. London, UK: Academic Press, 1977. Youn I, Fu FH, Suh J-K. Determination of the mechanical properties of articular cartilage using a high-frequency ultrasonic indentation technique. Trans Orthop Res Soc 1999;24:162. Zheng YP, Mak AF. An ultrasound indentation system for biomechanical properties assessment of soft tissues in-vivo. IEEE Trans Biomed Eng 1996;43:912–918.