| 110 | Katariina Nissinen | Katariina Nissinen Diffusion...

Publications of the University of Eastern Finland Dissertations in Forestry and Natural Sciences

Publications of the University of Eastern Finland

Dissertations in Forestry and Natural Sciences

isbn 978-952-61-1158-2

Katariina Nissinen

Diffusion of X-ray Contrast Agents in Intact and Repaired Articular Cartilage

Early degenerative changes

of articular cartilage are not

diagnosed sensitively enough with

conventional imaging techniques.

The diagnostic potential of contrast

enhanced computed tomography

(CECT) was investigated in this

thesis in vitro and in vivo. Imaging

the diffusion dynamics of the

contrast agent in articular cartilage

was found to provide valuable

information about the integrity and

composition of cartilage matrix.

Based on the present findings, it

seems that delayed quantitative CT

arthrography (dQCTA i.e. CECT in

vivo) is a potentially useful method

in the quantitative diagnostic

imaging of articular cartilage.

dissertatio

ns | 110 | K

atar

iina N

issinen

| Diffu

sion

of X-ra

y Co

ntra

st Agen

ts in In

tact a

nd

Rep

aired

Articu

lar C

artila

ge

Katariina NissinenDiffusion of X-ray Contrast

Agents in Intact and Repaired Articular Cartilage

KATARIINA NISSINEN NÉE KULMALA

Diffusion of X-raycontrast agents in intact

and repaired articularcartilage

Publications of the University of Eastern FinlandDissertations in Forestry and Natural Sciences

No 110

Academic DissertationTo be presented by permission of the Faculty of Science and Forestry for public

examination in the Auditorium in Mediteknia Building at the University of EasternFinland, Kuopio, on July, 31, 2013,

at 12 o’clock noon.

Department of Applied Physics

Kopijyvä Oy

Kuopio, 2013

Editors: Prof. Pertti Pasanen, Prof. Pekka Kilpeläinen,

Prof. Kai Peiponen, Prof. Matti Vornanen

Distribution:

University of Eastern Finland Library / Sales of publications

P.O. Box 107, FI-80101 Joensuu, Finland

tel. +358-50-3058396

http://www.uef.fi/kirjasto

ISBN: 978-952-61-1158-2 (printed)

ISSNL: 1798-5668

ISSN: 1798-5668

ISBN: 978-952-61-1159-9 (pdf)

ISSN: 1798-5676

Author’s address: University of Eastern FinlandDepartment of Applied PhysicsP.O.Box 162770211 KUOPIOFINLANDemail: [email protected]

Supervisors: Professor Juha Töyräs, PhDUniversity of Eastern FinlandDepartment of Applied PhysicsKUOPIO, FINLANDemail: [email protected]

Professor Jukka Jurvelin, PhDUniversity of Eastern FinlandDepartment of Applied PhysicsKUOPIO, FINLANDemail: [email protected]

Adjunct Professor Rami Korhonen, PhDUniversity of Eastern FinlandDepartment of Applied PhysicsKUOPIO, FINLANDemail: [email protected]

Reviewers: Adjunct Professor Harri Sievänen, PhDUKK InstituteTAMPERE, FINLANDemail: [email protected]

Professor Kunle Oloyede, PhDQueensland University of TechnologyScience and Engineering FacultyBRISBANE, QLD, AUSTRALIAemail: [email protected]

Opponent: Professor Christian Langton, PhDQueensland University of TechnologyScience and Engineering FacultyBRISBANE, QLD, AUSTRALIAemail: [email protected]

ABSTRACT

Early degenerative changes of articular cartilage are not diagnosedsensitively enough with conventional X-ray and magnetic resonanceimaging (MRI) techniques. Contrast agent enhanced MRI and com-puted tomography (CT) have also been applied for imaging degen-erative changes in cartilage. Contrast enhanced CT (CECT) can beconsidered as an analogous X-ray technique to delayed gadolinium-enhanced MRI of cartilage (dGEMRIC). Both techniques are based onimaging the diffusion of anionic contrast agent in the cartilage ma-trix. In this thesis, the diagnostic potential of CECT was investigatedin three in vitro studies and one in vivo study.

In study I, diffusion coefficients of four CT and MRI contrastagents in cartilage were determined by analyzing CECT measure-ments using the finite element method (FEM). In addition, diffusionthrough the articular surface was compared with diffusion throughthe deep cartilage. Study II evaluated the ability of CECT to dis-tinguish spontaneously healed osteochondral lesions from intactequine cartilage. In study III, the effects of increase of cross-linkingof collagen on diffusion of anionic and non-ionic contrast agentswere investigated using CECT. In study IV, dGEMRIC and delayedquantitative CT arthrography (dQCTA i.e. CECT in vivo) were com-pared with each other and to arthroscopic findings in a clinicalstudy.

Diffusion coefficients for four contrast agents (ioxaglate, gadopen-tetate, iodide, gadodiamide) could be determined in study I. Dif-fusion through superficial cartilage was faster than through deepcartilage. This appeared to reflect the differences in compositionalong cartilage depth: higher water content and lower glycosamino-glycan concentration in superficial than in deep cartilage. In study II,CECT was able to distinguish in a quantitative manner the sponta-neously healed cartilage and subchondral bone from adjacent intacttissues. In study III, cross-linking induced changes in cartilage fixedcharge density had greater effect on diffusion of anionic contrastagent as compared to the changes in the steric hindrance. In study

IV, the results of dQCTA normalized with the contrast agent concen-tration in synovial fluid showed the best agreement with dGEMRICat 45 minutes after contrast agent injection. However, neither ofthe techniques correlated with the arthroscopic evaluation. Sincenormalization with the contrast agent concentration in synovial fluidwas important, it was postulated to be taken into account in futureclinical application of dQCTA.

To conclude, imaging the diffusion dynamics of the contrastagent in articular cartilage was found to provide valuable informa-tion about the integrity and composition of cartilage matrix. Fur-thermore, CECT provided quantitative information on subchondralbone microstructure and density. Based on the in vivo experiment, itseems that dQCTA is a potentially useful method in the quantitativediagnostic imaging of articular cartilage.

National Library of Medicine Classification: QT 36, WE 300, WE 348,WN 160, WN 206PACS Classification: 87.19.xn, 87.57.-s, 87.57.Q-, 87.85.PqMedical Subject Headings: Cartilage, Articular; Collagen; Contrast Media;Diagnostic Imaging; Diffusion; Finite Element Analysis; Osteoarthritis;Tomography, X-Ray ComputedYleinen suomalainen asiasanasto: diffuusio; kollageenit; kuvantaminen;nivelrikko; nivelrusto; tietokonetomografia; varjoainetutkimus

To my dearestJani and Akseli

Preface

This thesis was carried out during the years 2008–2013 in the De-partment of Applied Physics, University of Eastern Finland.

I would like to express my gratitude to my supervisors for theirprofessional guidance during this thesis. In our weekly meetings,they have helped to find answers to various problems that I havemet on the way. Their expertise has significantly contributed tothe progress of this project. I would like to thank my principalsupervisor, Juha Töyräs, for his enthusiastic attitude to research. Iam grateful to my second supervisor, Jukka Jurvelin, for giving methe opportunity to work in his spirited research group, Biophysicsof Bone and Cartilage (BBC). I thank my third supervisor, RamiKorhonen, for his guidance especially at the beginning of this project,and for arranging the occasional ranch parties.

The official reviewers of this thesis, Harri Sievänen and KunleOloyede, are acknowledged for their professional review and con-structive criticism. Ewen MacDonald is thanked for the linguisticreview.

I would like to thank my co-authors for their significant con-tributions to the studies of this thesis. Particularly, I thank JukkaHirvasniemi for his efforts in the fourth study.

I owe my sincere thanks to the colleagues that I have had aprivilege to work with in BBC. We have shared so many fun momentsand supported each other during hard times. Janne Karjalainen andMarkus Malo are thanked for providing a cosy chair and a nicecup of tea in their office. We have shared rich conversations withErna Kaleva and Elli-Noora Salo while stitching and bitching. JukkaLiukkonen has proved to the former tech student that Vappu canalso be celebrated in Kuopio. Siru Turunen has been in the spirit allthe way from Germany. Antti Aula, Harri Kokkonen, and TuomoSilvast deserve special thanks for helping me to get started with the

research of cartilage imaging.My special thanks go to Emmi Laukkanen and Mika Mikkola

with whom I travelled between Tampere and Kuopio. Those journeyswere hilarious and made early Monday mornings much easier.

I would like to thank the staff of the Department of AppliedPhysics, Institute of Biomedicine, and SIB Labs for their help duringthis project. Especially, I thank Virpi Tiitu for her cheerful personalityand encouragement.

The following sources of funding are acknowledged: Emil Aal-tonen Foundation, Finnish Cultural Foundation, Finnish ConcordiaFund, Finnish Cultural Foundation - North Savo Regional Fund,Foundation of Advanced Technology of Eastern Finland, Interna-tional Graduate School in Biomedical Engineering and MedicalPhysics, Kuopio University Hospital (EVO grant 5041715), MagnusEhrnrooth Foundation, Päivikki and Sakari Sohlberg Foundation,and Sigrid Jusélius Foundation. Atria Suomi Oy is acknowledgedfor providing bovine knees for research. CSC – IT Center for Scienceis acknowledged for providing computing services.

I owe my deepest gratitude to my parents, Ritva and Seppo, andmy siblings, Anni and Konsta. Mom and dad, Anni and Konsta –you have always, always been there for me and encouraged me inwhat I do. Your support means a lot to me.

I have good memories from the years I spent in Kuopio, eventhough working there meant living apart from my love. Eventually,everything turned out fine. I wish to express my dearest thanks tomy beloved Jani. You helped me to spur myself to finish this project.Our little son Akseli, you are the light of my life.

Pirkkala, June 2013 Katariina Nissinen

ABBREVIATIONS

1D One-dimensional2D Two-dimensional3D Three-dimensionalAGE Advanced glycation endproductsC CartilageCA4+ Contrast agent carrying a positive charge of 4CCD Charge coupled deviceCECT Contrast enhanced computed tomographyCT Computed tomographyDD Digital densitometrydGEMRIC Delayed gadolinium-enhanced magnetic reso-

nance imaging of cartilageDMOAD Disease modifying osteoarthritis drugsdQCTA Delayed quantitative computed tomography

arthrographyECM Extracellular matrixEDTA Ethylenediaminetetraacetic acidFCD Fixed charge densityFE Finite elementFEA Finite element analysisFEM Finite element methodFLC Lateral condyle of femurFMC Medial condyle of femurFTIRI Fourier transform infrared imagingGAG GlycosaminoglycanHP HydroxylysylpyridinolineIA Intra-articularIV IntravenousICRS International Cartilage Repair SocietyLTG Lateral trochlear groove

LP LysylpyridinolinemicroCT X-ray microtomographyMR Magnetic resonanceMRI Magnetic resonance imagingMTG Medial trochlear grooveNEG Non-enzymatic glycationNMR Nuclear magnetic resonanceOA OsteoarthritisOCT Optical coherence tomographyPAT PatellaPBS Phosphate buffered salinePent PentosidinePET Positron emission tomographyPG ProteoglycanPI Parallelism indexPLM Polarized light microscopypQCT Peripheral quantitative computed tomographyROI Region of interestSD Standard deviationSF Synovial fluidTLC Lateral condyle of tibiaTMC Medical condyle of tibiaUS Ultrasonography

SYMBOLS AND NOTATIONS

A− Impermeable ionα Attenuation coefficientAn Anionb BathBMD Bone mineral densityBS Bone surfaceBV Bone volumeBS/BV Bone surface to volume ratioBV/TV Bone volume fractionc CartilageC ConcentrationCat CationCl− Chloride ionCO2 Carbon dioxideContrast Contrast agentD Diffusion coefficient∆R1 Change in relaxation timedGEMRIC-index T1,Gd, Longitudinal relaxation time in presence

of Gd-DTPA2−

e Euler’s numberE Young’s modulusF Faraday constantGd-DTPA2− Gadopentetate, gadolinium diethylenetriamine

pentaacetic acidΓ Diffusion fluxk PermeabilityK Partition coefficientK+ Potassium ionKCl Potassium chloriden Number of samples

Na+ Sodium ionν Poisson’s ratiop Measure of statistical significanceΨ Nernst potentialr Donnan ratioR Gas constantR Relaxivity or correlation coefficientt TimeT TemperatureT1,0 Longitudinal relaxation time in absence of Gd-

DTPA2−

T1,Gd Longitudinal relaxation time in presence of Gd-DTPA2−

TV Tissue volumex Position along the cartilage depthz Valency of an ion

LIST OF PUBLICATIONS

This thesis consists of the present review of the author’s work in thefield of contrast enhanced computed tomography of cartilage andthe following selection of the author’s publications:

I Kulmala KAM, Korhonen RK, Julkunen P, Jurvelin JS, QuinnTM, Kröger H and Töyräs J: “Diffusion coefficients of articularcartilage for different CT and MRI contrast agents”, MedicalEngineering and Physics 32, 878–82 (2010).

II Kulmala KAM, Pulkkinen HJ, Rieppo L, Tiitu V, Kiviranta I,Brünott A, Brommer H, van Weeren R, Brama PAJ, Mikkola MT,Korhonen RK, Jurvelin JS and Töyräs J: “Contrast-enhancedmicro-computed tomography in evaluation of spontaneousrepair of equine cartilage”, Cartilage 3, 235–44 (2012).

III Kulmala KAM, Karjalainen HM, Kokkonen HT, Tiitu V, Ko-vanen V, Lammi MJ, Jurvelin JS, Korhonen RK and Töyräs J:“Diffusion of ionic and non-ionic contrast agents in articularcartilage with increased cross-linking — contribution of stericand electrostatic effects”, Medical Engineering and Physics, Inpress (2013).

IV Hirvasniemi J*, Kulmala KAM*, Lammentausta E, Ojala R,Lehenkari P, Kamel A, Jurvelin JS, Töyräs J, Nieminen MTand Saarakkala S: “In vivo comparison of delayed gadolinium-enhanced MRI of cartilage and delayed quantitative CT arthrog-raphy in imaging of articular cartilage”, Osteoarthritis and Car-tilage 21, 434–42 (2013). (*Equal contribution)

Throughout the thesis, these publications will be referred to by theirRoman numerals. The original publications have been reprintedwith kind permission from the copyright holders (Appendices).

AUTHOR’S CONTRIBUTION

The publications selected in this dissertation are original research pa-pers on contrast enhanced computed tomography (CT) and delayedquantitative CT arthrography of cartilage.

In paper I, the author carried out the numerical analysis of theexisting experimental data and was the principal author of themanuscript.

In paper II, the author carried out the imaging experiments anddata analysis, participated in the finite element analysis and was theprincipal author of the manuscript.

In paper III, the author carried out the imaging experiments anddata analysis, and was the principal author of the manuscript.

In paper IV, the author carried out the CT data analysis, and wasthe principal author of the manuscript with an equal contributionwith J. Hirvasniemi.

In all papers, the co-operation with the co-authors has been signifi-cant.

Contents

1 INTRODUCTION 1

2 ARTICULAR CARTILAGE 52.1 Structure and composition . . . . . . . . . . . . . . . . 52.2 Functional properties . . . . . . . . . . . . . . . . . . . 82.3 Osteoarthritis . . . . . . . . . . . . . . . . . . . . . . . 9

2.3.1 Etiology . . . . . . . . . . . . . . . . . . . . . . 92.3.2 Diagnostic methods . . . . . . . . . . . . . . . 112.3.3 Treatment . . . . . . . . . . . . . . . . . . . . . 12

3 CONTRAST ENHANCED IMAGING OF CARTILAGE 153.1 Physics of contrast agent diffusion . . . . . . . . . . . 153.2 Delayed gadolinium enhanced magnetic resonance

imaging of cartilage . . . . . . . . . . . . . . . . . . . . 203.3 Contrast enhanced computed tomography . . . . . . 21

4 AIMS OF THE PRESENT STUDY 25

5 MATERIALS AND METHODS 275.1 Sample preparation . . . . . . . . . . . . . . . . . . . . 275.2 Contrast enhanced CT . . . . . . . . . . . . . . . . . . 29

5.2.1 Peripheral quantitative CT . . . . . . . . . . . 295.2.2 X-ray microtomography . . . . . . . . . . . . . 315.2.3 Delayed quantitative CT arthrography . . . . 32

5.3 Delayed gadolinium-enhanced MRI of cartilage . . . 335.4 Arthroscopy . . . . . . . . . . . . . . . . . . . . . . . . 335.5 Simulation of contrast agent diffusion . . . . . . . . . 345.6 Reference methods . . . . . . . . . . . . . . . . . . . . 35

5.6.1 Histological and biochemical analyses . . . . . 355.6.2 Mechanical testing . . . . . . . . . . . . . . . . 36

5.7 Statistical analyses . . . . . . . . . . . . . . . . . . . . 38

6 RESULTS 396.1 Contrast agent diffusion through superficial and deep

cartilage . . . . . . . . . . . . . . . . . . . . . . . . . . 396.2 CECT of spontaneously repaired osteochondral injuries 416.3 Effect of collagen cross-linking on contrast agent dif-

fusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . 426.4 dQCTA and dGEMRIC – association to arthroscopic

findings . . . . . . . . . . . . . . . . . . . . . . . . . . . 44

7 DISCUSSION 477.1 Contrast agent diffusion through superficial cartilage

is faster than through deep cartilage . . . . . . . . . . 477.2 CECT enables detection of repaired

osteochondral injuries . . . . . . . . . . . . . . . . . . 487.3 Cross-linking related changes in FCD have greater

effect on the contrast agent diffusion than changes insteric hindrance . . . . . . . . . . . . . . . . . . . . . . 49

7.4 dQCTA is in agreement with dGEMRIC in vivo . . . . 507.5 Limitations . . . . . . . . . . . . . . . . . . . . . . . . . 517.6 dQCTA – current status . . . . . . . . . . . . . . . . . 537.7 dQCTA – future . . . . . . . . . . . . . . . . . . . . . . 55

8 SUMMARY AND CONCLUSIONS 57

REFERENCES 59

APPENDICES (ORIGINAL PUBLICATIONS) 80

1 Introduction

Articular cartilage is a specialized connective tissue covering theends of the articulating bones, e.g. in the knee joint. Normal articularcartilage is a bluish white, hydrated soft tissue with a unique com-position and mechanical properties that enable it to withstand highloads and provide almost frictionless motion in the joint [117, 121].Cartilage and the menisci contribute to the distribution of loads inthe knee joint, and thus protect the articulating bones from dam-age [31].

The main components of cartilage are water and two structuralmacromolecules, collagen and proteoglycans (PGs). The highlyorganized collagen network and the negatively charged PGs forma porous solid matrix which is filled with interstitial water. Theinteractions between the solid matrix and water represent the basisof the biomechanical behaviour of cartilage under loading [117].Any disturbances in the main components affect the interactionsbetween them, and thus the function of cartilage is impaired. Theimpairment of the joint due to the presence of degenerative changesin cartilage and subchondral bone is called osteoarthritis (OA), oralternatively osteoarthrosis or degenerative joint disease [25, 31].

OA is the most common joint disease, which causes joint painand limited mobility [31]. The probability of having knee OA in-creases with age and is more common in women. Since OA oftenleads to total joint replacement and disability to work, the yearlycosts for society are high [68].

The earliest osteoarthritic changes occurring in cartilage aredisruption of the superficial collagen network, a decrease of PG con-centration and an increase in the water content [31, 111]. Changes inbone include thickening of the subchondral plate, increased densityof the trabecular bone underneath the subchondral plate and theformation of cysts and osteophytes [31]. Unfortunately, the earlysigns of OA are not visible in plain radiography which still is a

Katariina Nissinen: Diffusion of contrast agents in cartilage

common way to diagnose OA [35]. Magnetic resonance imaging(MRI) provides the possibility to image soft tissue, in contrast tonative X-ray techniques, but is more expensive and not available forall patients. Furthermore, the resolution of MRI is not high enoughto allow the detection of the incipient osteoarthritic changes or minorcartilage injuries.

During the past years, the application of contrast agents withMRI and computed tomography (CT) has shown potential in imag-ing compositional changes of articular cartilage. Delayed gadolinium-enhanced MRI of cartilage (dGEMRIC) was introduced in the mid1990s [16–18]. This method is based on the assumption that nega-tively charged contrast agents will distribute within cartilage in aninverse relation to the negative fixed charged density (FCD) asso-ciated with the glycosaminoglycans (GAGs) of PGs. Since the PGconcentration is lower in degenerated cartilage, more contrast agentis believed to diffuse into osteoarthritic tissue.

An analogous X-ray technique to dGEMRIC, contrast enhancedCT (CECT), was introduced in 2006 [37, 135]. Many in vitro studieshave revealed its potential to assess the state of cartilage [9, 13, 79,139, 163, 209]. However, diffusion and distribution of contrast agentin cartilage are complex phenomena which are also affected by otherconstituents in addition to PGs. The collagen concentration andintegrity of the collagen network, water content in cartilage, chargeand molecular weight of the diffusing molecule all contribute to dif-fusion [79,88,104,107,108,137,180,200,201]. Therefore, the specificityof both dGEMRIC and CECT as ways to measure the PG content hasbeen questioned [66, 152, 164, 171]. CECT has recently been tested invivo [86], but several questions remain open concerning the optimalamount of contrast agent and the optimal imaging time points inrelation to contrast agent injection. CECT is referred to as delayedquantitative CT arthrography (dQCTA) when it is used in in vivosettings.

In addition to early detection of OA, new sensitive imaging tech-niques of cartilage are needed for the development of medical andsurgical treatments [20, 26, 141]. This thesis focuses on the further

2 Dissertations in Forestry and Natural Sciences No 110

Introduction

development of contrast enhanced CT of cartilage, and consists ofthree in vitro studies and one in vivo study. Study I examined thetissue-averaged axial diffusion coefficients of four different CT andMRI contrast agents in cartilage. The ability of CECT to distinguishspontaneously healed osteochondral lesions from adjacent normalcartilage tissue was investigated in study II. Study III aimed to de-termine the individual contribution of increased FCD and sterichindrance of artificially aged, cross-linked collagen network in thepartition of contrast agents. Study IV was an in vivo study investigat-ing the performance of dGEMRIC and dQCTA and their associationto arthroscopic findings of cartilage status.

Dissertations in Forestry and Natural Sciences No 110 3

2 Articular cartilage

Articular cartilage (hyaline cartilage) is a specialized connectivetissue located at the ends of the articulating bones. The meancartilage thickness in a human knee joint is 2–4 mm depending onthe anatomical location [8, 64]. In a knee joint, cartilage efficientlydistributes loads together with menisci minimizing the stresseson the bones [29, 117]. Smooth cartilage surfaces provide almostfrictionless motion in the joints together with the lubricating synovialfluid. Cartilage tissue is characterized as a biphasic material withfluid and solid phases [117]. The main constituents of the solid phaseare collagen fibers and PGs. The fluid phase consists of interstitialwater and electrolytes. The fluid and solid phases together formthe extracellular matrix (ECM) which surrounds the cartilage cells,chondrocytes.

2.1 STRUCTURE AND COMPOSITION

Water accounts for approximately 60–80% of the wet weight ofarticular cartilage [122, 193]. Some of this water is bound to collagenfibrils but most of it can move freely in and out of the tissue. Theflow of water greatly affects the mechanical properties of the tissue.Water is also responsible for the transport of nutrients and ions.Since cartilage is avascular, the transport occurs through diffusionand convection from the synovial fluid. Tissue water contains ahigh concentration of cations to balance the negative FCD caused byPGs [29]. The water content decreases from the superficial cartilagetowards the cartilage–bone interface [157, 193].

Collagen accounts for about 10–20% of the wet weight of thecartilage [122] with type II collagen representing more than 90% ofthe total collagen [33]. Collagen fibers form a network which hasa unique, arcade-like organization [21]. Zonal differences in fiberarrangement and the extensive cross-linking of collagen confer the



Figure 2.1: Schematic illustration of the structure of articular cartilage (on the left) andproteoglycan aggregate (on the right). In the superficial zone (5–10% of cartilage thickness),the collagen fibers are oriented parallel to the surface. In the middle zone (5–20%), the fibersbend so that they start to become perpendicular to the surface. In the deep zone (70–90%),the fibers are oriented in a perpendicular manner to the surface and are anchored to thesubchondral bone through the calcified matrix. PG monomers are bound from one end to ahyaluronic acid chain to form an aggregate. A monomer is composed of a protein core withGAG chains attached to it.


Articular cartilage

tensile and shear stiffnesses and strengths on the tissue [50, 81, 117,120, 122, 211]. Cartilage is organized into three zones according tothe preferential direction of the fibers [21, 72]. In the superficialtangential zone (5–10% of cartilage thickness [6]), the fibers areoriented in parallel to the surface and are densely packed. In themiddle zone (5–20% [6]), the fibers bend so that they start to becomeperpendicular to the surface. In the deep zone (70–90% [6]), thefibers are oriented in a perpendicular manner to the surface and areanchored to the bone through the calcified cartilage (Figure 2.1).

Collagen fibers are extensively cross-linked which is necessaryfor the high tensile strength of the healthy collagen network [11, 50].Some of the cross-links are formed through non-enzymatic glycation(NEG) which is a typical phenomenon in natural ageing. NEG isa reaction between reducing sugars and proteins which produceadvanced glycation endproducts (AGEs). AGEs accumulate in allproteins, but they are especially prevalent in collagen due to its slowturnover rate [195]. Cartilage tissue can be artificially aged in vitroby creating cross-links via sugar-induced NEG. Threose (sugar) hasbeen used to increase collagen cross-linking in vitro and this hasresulted in an increase of the stiffness of the collagen network [194].

PGs account for 5–10% of the wet weight of the cartilage [117]. APG monomer is composed of a protein core with chains of GAG, suchas chondroitin sulphate and keratan sulphate, covalently attached toit [122]. PG monomers are bound from one end to hyaluronic acidchains to form large aggregates (Figure 2.1) [33, 119]. Negativelycharged GAGs create a fixed negative charge within the cartilagematrix [109], and thus attract cations and water into the cartilage.This together with the repulsive forces between the GAG chains givesrise to the osmotic swelling pressure which contributes primarily tothe compressive stiffness of cartilage [121, 122]. Swelling pressure,which generates a prestress in the collagen network, contributesalso to the shear stiffness of cartilage [211]. In humans, the PGconcentration increases from the superficial cartilage to the deepzone [124, 193].

Chondrocytes are specialized cells; their main function is to



produce and maintain the ECM of cartilage. The cells accountfor only about 1% of the volume of the human cartilage [29, 70,71]. The chondrocytes are more active in young than in matureindividuals. The activity of chondrocytes to synthesize collagenand PG in order to replace the degraded matrix components isinfluenced by their mechanical and physicochemical environment[29, 70, 123]. Nevertheless, the chondrocytes have a limited capacityto maintain the cartilage, especially after mechanical injury. Thecauses behind the decline in chondrocytic metabolic response arenot fully understood [29, 121].

2.2 FUNCTIONAL PROPERTIES

PGs are immobilized and entrapped within the fine network ofthe collagen fibers. Since the GAG chains are very close to eachother, the repulsive forces between them cause the ECM to swell(the chemical expansion effect). In addition, the FCD created bynegatively charged GAGs causes the cation concentration insidethe tissue to rise, leading to an osmotic pressure difference andswelling of the tissue (Donnan osmotic pressure effect). The collagennetwork is prestressed by the swelling, and it resists the deformationof the matrix. Thus, the swelling behaviour and the anisotropicviscoelastic collagen network can explain how cartilage responds toloading. [96, 110, 117, 205]

A viscoelastic material responds to loading in a time-dependentmanner [117]. When the cartilage is under the compressive load, thelow permeability helps the tissue to resist the fluid flow within andout of the tissue. Therefore, the pressurized fluid can support asmuch as 95% of the applied load [7, 118, 133]. At the same time, thecollagen tension resists the tissue deformation in a strain-dependentmanner [81, 115, 149]. Thus, collagens and pressurized fluid mainlysupport the load in dynamic compression. Under a constant load,the fluid flows slowly through the porous tissue and this governs thetransient response of cartilage. At equilibrium (static loading), PGsare mostly responsible for the load support [91, 95, 117]. After the


Articular cartilage

release of the load, fluid flows back into cartilage and it swells backto its original volume. Since the fluid flow has an important role inthe mechanical behaviour of the cartilage, poroelasticity describesbetter the behaviour of cartilage than viscoelasticity [113].

Computational biomechanical models of articular cartilage canbe exploited to better understand the biomechanical behaviour andproperties of articular cartilage [76,207]. The biphasic mixture modelof cartilage, which separates fluid and solid phases, was introducedin 1980 [120]. It is essentially analogous to the poroelastic theoryintroduced in 1941 [22]. In 1991, a triphasic mixture model wasintroduced, which includes solid, fluid, and ionic phases [96]. Carti-lage has also been modeled as a poroviscoelastic [172], transverselyisotropic [38], and conewise linear elastic [166] material. In the late1990’s the first fibril-reinforced model including the collagen networkand poroelastic matrix (PGs and fluid) was developed [103]. Later,fibril-reinforced poroelastic model was extended to fibril-reinforcedporoviscoelastic [206] and fibril-reinforced poroviscoelastic swellingmodels [205].

2.3 OSTEOARTHRITIS

Osteoarthritis (OA) is the most common disease in synovial joints,most often found in the knee, hip, spine, foot and hand joints [31].It is characterized by pain, stiffness and decreased mobility, andthus diminishes the ability to function and the quality of life of apatient. Both men and women suffer from OA, which becomes morecommon with age [126]. OA is a costly disease for society, not onlybecause of the expensive surgical treatments, but also due to theneed for OA-based disability pensions [68].

2.3.1 Etiology

The reasons that lead to the development of OA are not fully under-stood [32]. In OA, a progressive loss of articular cartilage is accom-panied by attempted repair of the cartilage, remodeling and sclerosis



of subchondral bone, and the formation of bone cysts and marginalosteophytes [30]. Most commonly, OA develops with increasing agewithout any clear cause. This is referred to as primary OA. Instead,secondary OA develops as a result of an injury, infection or somedevelopmental or hereditary reasons such as an abnormal shape ofa joint [30]. In addition to age, obesity and excessive mechanicalloading are important risk factors of OA [32, 51, 158].

The earliest degenerative changes in articular cartilage includeloss of PGs and an increase in water concentration [31, 111]. Alter-ations in the collagen network cause swelling of the tissue [30, 31].These changes increase the permeability and decrease the stiffnessof cartilage, which in turn can accelerate the rate of degeneration.Visual changes of the cartilage include fibrillation of the superficialtissue and thinning of cartilage which eventually leads to a loss ofcartilage, exposing the underlying subchondral bone [31]. Some ofthe aforementioned degenerative changes are difficult to separatefrom normal age-related changes in articular cartilage [24, 111].

Chondrocytes detect the degenerative changes in cartilage, and tosome extent can stabilize or even restore the tissue [30,31]. However,when the degradative processes overwhelm the reparative efforts,chondrocytes fail to protect the tissue from further damage [121].

Alterations in the subchondral bone include thickening of sub-chondral plate, increased density of the subchondral trabecularbone, formation of cysts, and appearance of subchondral carti-lage [30, 31, 59]. According to some speculations OA starts in sub-chondral bone in a way that stiffening of the bone causes increasedloading of cartilage [53, 59, 142]. Bone cells (i.e. osteoblasts andosteoclasts [169]) react more rapidly and effectively than chondro-cytes to changes in the biomechanical environment by remodelingand modeling. This difference in adaptive capacity of bone andcartilage affects the physiological relationship between these tissuescontributing to the further development of OA [59].


Articular cartilage

2.3.2 Diagnostic methods

Diagnosis of OA is based on the symptoms of the patient andfindings in clinical examination, such as palpable effusion or visiblemalalignment of the joint [4, 99]. In radiographically defined OAnarrowing of the joint space, increased density of subchondral boneor presence of osteophytes are visible in plain radiographs [32].Patients having both clinical symptoms and radiographic evidenceof OA are considered to have symptomatic OA [99]. In the followingsections, conventional methods to diagnose OA are reviewed andnew techniques still under development are introduced.

Radiography is a conventional way to diagnose OA. The Kellgren-Lawrence grading system is widely used to evaluate the severity ofOA from plain radiographs [80]. In this system, the formation ofosteophytes, narrowing of joint space, sclerosis of subchondral boneand possible changes in the shape of the bone ends are evaluated[80]. Unfortunately, these changes are signs of OA that has alreadyextensively developed and little can be done to save the joint fromfurther damage. Furthermore, the symptoms of the patient andthe radiographic findings do not always correlate [32, 112]. Eventhough CT provides three-dimensional (3D) imaging of the kneejoint, it suffers from the same disadvantages as plain radiography:its inability to image cartilage and to detect early changes in OA [35].

Arthroscopy is an invasive method which allows a visual in-spection of the cartilage surfaces. It is not generally used for thediagnostics of OA but for the evaluation of cartilage lesions it servesas the gold standard [168]. Cartilage lesions can be classified indifferent ways e.g. according to the International Cartilage RepairSociety scoring system which focuses on the lesion depth and thearea of damage [28]. However, this grading system has been re-ported to suffer from a high degree of inter- and intraobservervariability [167, 168].

MRI with properly chosen imaging sequences enables imagingof cartilage morphology, and can detect changes in subchondralbone and meniscal abnormalities [41, 67]. However, its rather poor



spatial resolution, high costs and long acquisition times limit its usein the diagnostics of cartilage injuries and degeneration. Quantita-tive MRI techniques to assess the composition of cartilage are alsoavailable [41]. One of these is delayed gadolinium-enhanced MRI ofcartilage (dGEMRIC) which has been proposed to provide a quanti-tative estimation of the GAG concentration in cartilage [18, 176, 202].dGEMRIC is based on the assumption that the anionic gadopentetate(Gd-DTPA2−) molecule is distributed in cartilage, at equilibrium,in an inverse manner to the GAG concentration. However, it hasrecently been reported that the equilibrium partitioning is not achiev-able in vivo, meaning that the technique is sensitive to other factorsaffecting diffusion, e.g. mechanical injury [66, 87, 152, 171].

Ultrasonography (US) of the joints is a non-invasive low-costimaging method that enables visualization of articular surfaces andthe contours of the subchondral bone. Thinning of cartilage, thick-ening of the synovial membrane and effusion of the joint can alsobe detected by US. The main limitations of the technique are its in-ability to image some of the weight bearing areas due to shadowingby surrounding tissues and its low resolution, not enabling detec-tion of initial degenerative changes in cartilage and subchondralbone. [1, 116, 151]

Ultrasound arthroscopy [69, 77, 196] and arthroscopic opticalcoherence tomography (OCT) [36, 210] have recently been proposedfor quantitative high resolution imaging of articular cartilage.

2.3.3 Treatment

Currently, there is no treatment effective enough to stop the progres-sion of OA. However, there are some treatments that can be appliedto improve the quality of life of patients, especially if started earlyenough. Treatment options can be divided into nonpharmacologic,pharmacologic and surgical treatment. [31]

Since obesity is a major risk factor of OA, weight loss can helpto reduce symptoms and may even slow down the progression ofthe disease. The exercise program should include range-of-motion


Articular cartilage

and stretching exercises, muscle strengthening exercises and aerobicexercises, but activities involving high loading of the joints shouldbe avoided [31]. Suitable physical exercise and reduction of obesityare the most efficient ways to prevent OA [52].

Pharmacologic therapies include analgesics, nonsteroidal anti-inflammatory drugs and intra-articular injections of steroids andhyaluronans [141]. Generally, these medications relieve pain buthave virtually no effect to stop the structural degradation of cartilage.New drugs called disease modifying osteoarthritis drugs (DMOADs)are under development [141].

There are several surgical procedures that attempt to repair carti-lage injuries such as bone marrow stimulation [140, 146], autologouschondrocyte transplantation [27] and autologous osteochondral mo-saicplasty [65]. Bone marrow stimulation, i.e. microfracturing, isthe most frequently used technique. The technique involves makingmultiple holes in the subchondral bone plate through which thebone cells could enter from the bone marrow into the cartilage lesion.These cells are able to differentiate into fibrochondrocytes. However,the subsequent repaired fibrocartilage does not correspond to thesurrounding hyaline cartilage and has less type II collagen. Thehealing results have also been reported to deteriorate over time [20].

Autologous chondrocyte transplantation involves at least twooperations, one for tissue harvest and the second for cell transplanta-tion. First, cartilage slices are obtained from the less weight bearinglocations of the joint of the patient. Then, the cells are isolated andcultured and finally injected into the lesion under a periosteal flapwhich has been sutured to cartilage to cover the lesion [27]. With thistechnique, hyaline cartilage can be restored at the repair site [20].

Autologous osteochondral mosaicplasty involves obtaining mul-tiple small cylindrical osteochondral grafts from the less weightbearing locations of the joint of the patient and transplanting themto the cartilage lesion [65]. The procedure can be performed in asingle operation. Transplantation of living hyaline cartilage has ledto encouraging outcomes, eventhough there may be differences inorientation, thickness and mechanical properties between the donor



and recipient cartilages [20].Total knee replacement is a safe and cost-effective treatment for

patients suffering from constant pain not relieved by non-surgicaltreatment and with significantly impaired function [127]. About 85%of patients are satisfied with the results of surgery, but sometimes arevision operation is needed [127]. According to recommendationsby the Finnish Medical Society Duodecim for the management ofOA, total knee replacement should only be considered as the finaloption [47].


3 Contrast enhanced imagingof cartilage

The ability of X-ray imaging to differentiate between tissues dependon differences in their capabilities to attenuate X-rays. Attenuationdepends on the density and atomic number of the tissue in question,and the energy of the X-rays in use. In a knee joint, there is no naturalcontrast between the synovial fluid, cartilage and menisci. Contrastagents with high atomic numbers can be used to highlight specificstructures in the knee joint. Mostly iodine-based agents are usedin X-ray imaging, due to their solubility and low toxicity [148]. InMR-imaging, the paramagnetic contrast agent, usually gadolinium-based, alters the relaxation times of hydrogen atoms within thetissue, and thus improves contrast. Gadolinium-based contrastagents are safe to use provided that the patient does not suffer fromkidney disease [188].

3.1 PHYSICS OF CONTRAST AGENT DIFFUSION

Since cartilage is avascular, transport of contrast agents into cartilageoccurs through diffusion. Diffusion is caused by random movementof the molecules, e.g. ions. If there is a concentration gradient,diffusion occurs from a higher concentration to a lower one leadingto complete mixing [42]. Before applying the diffusion theory tocartilage, the basic physics behind equilibrium of ions in a situationwhere some ions are not able to diffuse through a semi-permeablemembrane, will be reviewed. The equilibrium in the aforementionedsituation is known as the Donnan equilibrium [45, 58].

Let us consider a bowl with two chambers separated by a semi-permeable membrane. On side 1, we have a solution containingpermeable potassium ions K+ and impermeable ions, A−. On side



2, we have potassium chloride (KCl) with permeable ions K+ andCl−. At the beginning, the concentrations of the ions are C1 and C2

on sides 1 and 2, respectively.

[K+]1 = [A−]1 = C1 and [K+]2 = [Cl−]2 = C2

Since there are no Cl− on side 1, the concentration gradient forcesCl− to diffuse from side 2 to side 1.

[Cl−]1 = x and [Cl−]2 = C2 − x

This results in a higher negative charge on side 1, which leads tothe presence of an electrical gradient across the membrane. As aconsequence, K+ starts to diffuse from side 2 to side 1 to balance theelectrical unequilibrium.

[K+]1 = C1 + y and [K+]2 = C2 − y

This, on the other hand, induces a concentration gradient of K+.At equilibrium, there is no netflow through the membrane, but

the mobile ions, K+ and Cl− continue to diffuse back and forth inorder to balance the concentration and electrical gradients across themembrane. At equilibrium, both sides are electrically neutral them-selves, having equal concentrations of cations and anions. Becauseof the impermeable anions on side 1, there is a higher concentrationof negative ions on that side and this creates an electrical potentialdifference across the membrane. This is called the Nernst potentialwhich is equal for all mobile ions. Nernst equation is

∆Ψ = −RTzF

lnC1

C2, (3.1)

where R is the gas constant, T is temperature, z is the valency of theion, F is the Faraday constant, and C1 and C2 the concentrations ofthe ions on sides 1 and 2 of the membrane. By reorganizing equation(3.1), we have (

C1

C2

) 1z

= e−F∆Ψ

RT= r, (3.2)


Contrast enhanced imaging of cartilage

which is known as the Donnan ratio.At equilibrium in the above example, according to equation (3.2)

it can be seen that

[K+]1[K+]2

=[Cl−]2[Cl−]1

(3.3)

C1 + yC2 − y

=C2 − x

x. (3.4)

To maintain the electroneutrality, it must hold that x = y. Thus, ifwe solve equation (3.4) for x we have

x =C2

22C2 + C1

,

and we can write the Donnan ratio for the above example as follows.

r =[K+]1[K+]2

=[Cl−]2[Cl−]1

=C1

C2+ 1 (3.5)

As demonstrated with the previous example, a fixed charge onone side of the membrane forces the mobile anions and cations tobecome unevenly distributed on both sides of the membrane.

If we apply this theory to cartilage tissue, with mobile anions(An) and cations (Cat) and negative FCD, equilibrated in a bathcontaining electrolyte solution with anionic contrast agent (Contrast),the electrochemical equilibrium requires that(

[An]c[An]b

) 1zanion

=

([Cat]c[Cat]b

) 1zcation

=

([Contrast]c[Contrast]b

) 1zcontrast

, (3.6)

where z is valency of the ion and subscripts c and b refer to cartilageand bath, respectively.

In order to satisfy the electroneutrality in the bath and in thecartilage, the following conditions must be met:

zanion[An]b + zcation[Cat]b = 0 (3.7)

zanion[An]c + zcation[Cat]c + FCD = 0. (3.8)



It must be noted that in equations (3.7) and (3.8) it is assumedthat the concentration of contrast is negligible as compared to theconcentrations of the mobile anions and cations. By combiningequations (3.6) – (3.8) it is theoretically possible to calculate FCDfrom the measured contrast agent concentration and known bathionic concentration.

FCD =− zcation

[Cat]1

zcationb [Contrast]

1zcontrastc

[Contrast]1

zcontrastb

1

zcation

− zanion

[Cat]1

zanionb [Contrast]

1zcontrastc

[Contrast]1

zcontrastb

1

zanion

(3.9)

The application of dGEMRIC, which is presented later in this chapter,is based on reaching the ideal Donnan equilibrium between cartilagetissue and external fluid [17].

The actual transport via diffusion can be described with Fick’slaws of diffusion [42]. The Fick’s first law describes steady-statediffusion, where the solute flux Γ, i.e the number of particles persecond per unit tissue area, is related to the gradient of concentrationC with the diffusion coefficient D:

Γ = −D∂C∂x

, (3.10)

where C is solute concentration per volume of tissue and x is theposition along the cartilage depth. D is a specific constant for thesolute of interest and the tissue through which it is being transported,and it determines how quickly an equilibrium concentration can beachieved in the system [39].

In an unsteady situation, where the concentration gradient atthe diffusion interface is suddenly increased and a time-dependentconcentration profile is produced while the solute penetrates intothe tissue, the Fick’s second law is applicable [42].

∂C∂t

= −∂Γ

∂x= −D

∂2C∂x2 , (3.11)



Figure 3.1: Cartilage sample equilibrating in a contrast agent bath. Boundary 1 is adiffusion interface, and boundaries 2–4 are closed with a sample holder.

where t is time and D is assumed to be constant. Considering thesituation with a cartilage sample equilibrating in a bath containingcontrast agent (Figure 3.1), the following boundary conditions apply:

t = 0 [Contrast]c = 0 in the whole sample

t > 0 [Contrast]c = K[Contrast]b for boundary 1

Γ = 0 for boundaries 2, 3 and 4

where t is time and K is the partition coefficient assumed to beconstant as well as the contrast agent concentration in the bath([Contrast]b). Boundaries 2, 3 and 4 are closed with a special sampleholder. Given these boundary conditions, it is possible to solve equa-tion (3.11) using finite element analysis (FEA) which is a convenienttool to numerically solve partial differential equations. By fittingthe model to the concentrations measured e.g. via CT imaging, it ispossible to determine D.

Diffusion and partition of solutes in cartilage depend on manyfactors. As can be seen from equation (3.6), the higher the chargeof a particle, the higher the concentration gradient. Indeed, many



studies have reported the charge dependency of partitions of ionicand non-ionic contrast agents in cartilage [37, 57, 201]. In addition,the increasing size of the diffusing molecule has shown a negativecorrelation with D in studies with glucose, inulin and dextran [3,100,101,107,179–182]. The local organization of collagen network has alsoan effect on diffusion, resulting in the anisotropic D [101]. The FCDof cartilage is an evident factor especially controlling the diffusionof charged solutes, and it correlates negatively with D [106–108].FCD is also an important determinant of the pore size of the ECM[108]. Fine pore size due to high PG concentration causes sterichindrance, and especially large molecules (charged or not) can betotally excluded from regions with high PG concentrations [107,108].Variations in the water content are also related to the pore size ofthe ECM, and thus increase in water content correlates positivelywith D [106, 108].

3.2 DELAYED GADOLINIUM ENHANCED MAGNETICRESONANCE IMAGING OF CARTILAGE

In delayed gadolinium enhanced MRI of cartilage (dGEMRIC) dif-fusion of paramagnetic contrast agent gadopentetate (Gd-DTPA2−)into cartilage is imaged by determining its effect on tissue T1 relax-ation time. The basic idea is that the anionic contrast agent willdistribute in cartilage according to the ideal Donnan equilibrium [17],i.e. in an inverse relation to the distribution of negative FCD inducedby GAGs. Thus, assuming that there is a Donnan equilibrium be-tween cartilage tissue and external fluid, it is theoretically possibleto calculate FCD in cartilage according to equation (3.9).

Making the following annotations

Cat = Na+ zcation = 1

An = Cl− zanion = −1

Contrast = Gd-DTPA2− zcontrast = −2

and adding a corrective factor of 2 [18] we end up with the following



equation for FCD in cartilage:

FCD = 2[Na+]b

(√[Gd-DTPA2−]c

[Gd-DTPA2−]b−

√[Gd-DTPA2−]b

[Gd-DTPA2−]c

), (3.12)

where subscripts c and b refer to cartilage and bath, respectively.The concentration of sodium may be measured by nuclear magneticresonance (NMR) spectroscopy [102]. The concentrations of Gd-DTPA2− can be determined as follows:

[Gd-DTPA2−] =1R

(1

T1,Gd− 1

T1,0

), (3.13)

where R is the relaxivity, T1,Gd and T1,0 are the relaxation timesmeasured in presence and absence of Gd-DTPA2−, respectively. Incurrent practice, the T1,Gd, referred to as dGEMRIC-index, or changein relaxation rate (∆R1, expression in parentheses in equation (3.13))is reported.

dGEMRIC has been widely studied in vitro [2, 10, 17, 18, 84, 94,98, 129–131, 153, 185, 197, 208] and in vivo [16, 56, 93, 114, 125, 128,132, 134, 159, 173–178, 183, 184, 198, 202–204]. The in vitro resultshave revealed high correlations between the GAG concentrationand T1,Gd [17, 18, 84, 129, 131, 197, 208]. However, the validity of thetechnique has been recently questioned [66, 152, 171] if one uses itto try to achieve a quantitation of cartilage GAG content in vivo.The challenge with dGEMRIC is that the equilibrium partitioningmay not be reached in vivo due to slower diffusion of Gd-DTPA2−

into cartilage than had been assumed [152, 164] and partly dueto washout of the contrast agent before an equilibrium has beenreached in deep cartilage [66].

3.3 CONTRAST ENHANCED COMPUTED TOMOGRAPHY

Contrast enhanced computed tomography (CECT) is an X-ray tech-nique analogous to dGEMRIC [37,135]. Similarly as with dGEMRIC,there are several successful in vitro studies where the relationship be-tween anionic contrast agent concentration and GAG concentration



in cartilage has been shown [13,37,79,135,163–165,209]. The contrastagent concentration measured with CECT has also been shown tocorrelate with the cartilage biomechanical properties [9, 13]. How-ever, it has been reported that the diffusion time for contrast agentto reach the equilibrium may well be several hours, which suggeststhat it cannot be reached in clinical settings [79, 163–165, 199].

According to an earlier study [164], GAG and contrast agentconcentrations have correlated already before the equilibrium hasbeen reached. In addition, CECT has enabled the detection ofcartilage injuries already after 30–60 minutes of diffusion [87]. Theseresults imply that imaging of diffusion dynamics instead of thesituation at equilibrium may enable evaluation of cartilage integrity.

Recently, a new cationic iodinated contrast agent (CA4+) hasbeen introduced to be used in CECT [75] and it has been evaluatedin vitro [14, 15, 97] and in vivo in rabbits [170]. Carrying a positivecharge of 4 it is effectively attracted by the GAGs and accumulatesin higher concentrations in cartilage compared to anionic contrastagents and this makes easier the segmentation of cartilage [14, 15, 75,170]. CA4+ has been shown to provide a better correlation betweencontrast agent and GAG concentrations than anionic contrast agents[14, 15]. Similar to the situation of CECT with anionic contrastagents, CECT with CA4+ has also been reported to correlate withthe biomechanical properties of cartilage [97].

The capability of CECT to detect cartilage degeneration has beeninvestigated in vivo in small animal models, and the method hasbeen able to distinguish between healthy and degenerated carti-lage [138, 139, 161]. These animal studies and ex vivo studies oncadaveric knee joints have reported encouraging results showingthat CECT provides valuable information on cartilage status evenwithout reaching the diffusion equilibrium [160, 189].

X-ray arthrography has been in clinical use for decades [23, 105,136, 143]. Contrast agents have been used to visualize cartilagemorphology and superficial lesions, and to assess the menisci andligaments [5, 44, 55, 147, 190]. In contrast to arthrography, a delaybetween the contrast agent injection and image acquisition is ap-



plied in delayed quantitative CT arthrography (dQCTA i.e. in vivoapplication of CECT). The clinical potential of this technique hasbeen recently investigated for the first time [86]. In that study, thecontrast agent concentration in cartilage that was normalized withthe concentration in the synovial fluid was higher in an OA kneethan in a healthy knee at 30 minutes after intra-articular injection ofanionic contrast agent. Since each patient has a different amount ofsynovial fluid which dilutes the injected contrast agent, it is difficultto decide on the optimal amount and concentration of the contrastagent in order to provide the best possible diagnostic quality.

In addition, the minimum number of acquisitions needed indQCTA is still unclear. Three acquisitions were proposed by Kokko-nen et al. [86]: one before contrast agent injection (non-contrastimage), one after the injection followed by a few minutes of lightexercise of the joint to aid distribution of the contrast agent (arthro-graphic image), and one at 30–60 minutes after the injection (delayedimage). By subtracting the non-contrast image from the arthro-graphic and delayed images, it is possible to determine the absolutecontrast agent concentration in cartilage in these two time points.However, considering the radiation dose, acquisition of the non-contrast image may not be always acceptable. It is possible thatthe amount of contrast agent penetrating to cartilage between theacquisitions of the arthrographic and delayed images could serve asan adequate indicator of the state of the cartilage (Figure 3.2) [85].



Figure 3.2: Examples of dQCTA images of human tibiofemoral joint [85]: A) arthrographicimage with clearly visible superficial lesions, B) delayed image acquired at 45 min aftercontrast agent injection, and C) subtraction image which is acquired by subtracting thearthrographic image from the delayed image. The high contrast agent concentrationvisualized in the figures B and C is indicative of poor cartilage quality.


4 Aims of the present study

In this thesis the diffusion of X-ray contrast agents in cartilage wasinvestigated in vitro and in vivo. The aim was to assess the effects ofcartilage integrity and composition on contrast agent diffusion andto investigate the diagnostic potential of contrast enhanced CT.

The specific aims of this thesis were

1. to determine the tissue-averaged axial diffusion coefficientsfor four different CT and MRI contrast agents and to comparediffusion through the articular surface with diffusion throughdeep cartilage.

2. to investigate the ability of microCECT to distinguish sponta-neously healed osteochondral lesions from normal cartilagetissue in equine intercarpal joint.

3. to determine the individual contribution of cross-linking in-duced increases in FCD and steric hindrance to the change inthe partition of clinical X-ray contrast agents in cartilage.

4. to compare dGEMRIC and dQCTA with each other, and theirassociation to the arthroscopic findings in human knee jointsin vivo.


5 Materials and methods

This thesis consists of four independent studies (I–IV). The materialsand methods used in the studies are summarized in Table 5.1. Theexperimental CECT data in study I was obtained from an earlierstudy by Silvast et al. [165]. The rest of the data is original.

Table 5.1: Summary of materials and methods used in studies I–IV.

Study Species and tissue Number ofsamples orpatients

Methods

I Bovine cartilage n = 6 per group4 groups

CECTFE analysis of diffusion

II Horse cartilage andsubchondral bone

n = 7 microCECTFE analysis of diffusionMechanical testingFE analysis of cartilagedeformation under loadHistological analysis

III Bovine cartilage n = 6 in group In = 7 in group II

microCECTBiochemical analysisHistological analysis

IV Human cartilagein vivo

n = 11 dQCTdGEMRICArthroscopy

5.1 SAMPLE PREPARATION

The bovine knees for studies I and III were acquired from a localabattoir (Atria Oyj, Kuopio, Finland). The samples were detached



from upper lateral quadrant of visually intact patellae. In studyI, eight adjacent full-thickness (1.3 ± 0.1 mm, mean ± SD) cartilagediscs (diameter 4 mm) were detached from patellae and randomlydivided into four groups according to the contrast agent to be usedin CECT imaging [165].

In study II, osteochondral lesions (diameter 6 mm) were sur-gically created in the intercarpal joints of horses under generalanesthesia. After 12 months of spontaneous healing, the horses weresacrificed, and osteochondral plugs (diameter 14 mm) were har-vested. The plugs included the repair cartilage and intact adjacenttissue. The procedures were approved by the Utrecht UniversityAnimal Experiments Committee.

In study III, cylindrical plugs (diameter 24.5 mm) were drilledfrom the upper lateral quadrant of bovine patellae and cut intofour pieces. Osteochondral samples (diameter 6 mm) were punchedfrom each piece, and divided into two groups according to thecontrast agent to be used in CECT imaging. One of the pairedsamples served as a reference while the other was treated withthreose (Sigma Aldrich Co., St. Louis, MO, USA) to induce collagencross-linking, which resembles the process of natural ageing. Thesamples were incubated in humidified 5%CO2/95% air atmosphereat 37◦C for seven days.

Study IV involved contrast enhanced CT and MR imaging ofhuman patients (n = 11) referred to an arthroscopic surgery of theknee. One patient declined to undergo arthroscopy but completedall imaging studies. One patient was excluded from the analysis dueto irregular distribution of contrast agent in the joint. An informedconsent was obtained from all patients. The study was approvedby the Ethical Committee of the Northern Ostrobothnia HospitalDistrict, Oulu, Finland (No. 33/2010).

The rather low number of samples in the studies was acceptedfor practical reasons. However, it was ensured that in the in vitrostudies the control and experimental samples were paired, and thetest groups included independent sample pairs.


Materials and methods

5.2 CONTRAST ENHANCED CT

Contrast enhanced imaging experiments were conducted with threedifferent scanners. A peripheral quantitative CT (pQCT) scannerwas used in study I, microCT was used in studies II and III, and aclinical 64-slice CT scanner was used in study IV (Table 5.2).

Table 5.2: Imaging scanners and parameters used in studies I–IV.

Scanner Voxel size Tubevoltage orfieldstrength

Imaging time points

pQCT 200 × 200 µm2 58 kV 1, 5, 9, 16, 25, 29 h a

slice thickness: 2.3 mmmicroCT 29.8 × 29.8 × 29.8 µm3 100 kV Study II: Continuously

for 2.5 h, 3, 4, 5, 6, 7, 8, 9,10, 11, 12, 24, 26, 28, 30,32, 34, 36 h a

Study III: 5 min, 24 h a

CT 312 × 312 × 312 µm3 100 kV 5, 45 minMR 469 × 469 µm2 3 T 90 min a

slice thickness: 3.0 mmaPre-contrast image was acquired.

5.2.1 Peripheral quantitative CT

A clinical pQCT scanner (XCT 2000, StraTec Medizintechnik GmbH,Pforzheim, Germany) was used in study I. Tube voltage of 58.0 kV,slice thickness of 2.3 mm and in-plane pixel size of 200 × 200 µm2

were applied. Diffusion of four different contrast agents (ioxaglate,gadopentetate, iodide, gadodiamide, Table 5.3) was investigated for29 h. One of the paired cartilage discs was mounted in the sampleholder in a way that diffusion was allowed through the articularsurface, and the other disc in a way that diffusion was allowedthrough the deep cartilage. The samples were first imaged withoutcontrast agent, and after immersion in the contrast agent bath atthe following time points: 1, 5, 9, 16, 25 and 29 h. The final data



Table5.3:C

haracteristicsofthe

contrastagentsused

inthe

studiesI–IV.C

ontrastagentsw

erediluted

inphosphate

bufferedsaline

containinginhibitors

ofmetalloproteinases.In

studyIV,the

contrastagentsw

erediluted

inphysiologicalsaline,exceptfor

theintravenously

injectedfull

strengthM

agnevist TM

.

Contrast

agentA

ctivem

oietyC

hargeM

olarm

ass(g/m

ol)C

oncentration(m

M)

Manufacturer

Hexabrix

TM

Ioxaglate−

11269

StudyI:21

Mallinckrodt,St.Louis,M

O,U

SAStudy

II:10Study

III:21Study

IV:105

Guerbet,R

oissy,FranceM

agnevistR©

Gadopentetate

−2

548Study

I:100Bayer

ScheringPharm

a,Berlin,G

ermany

Magnevist T

MStudy

IV:2.5

and500

aBayer

HealthC

arePharm

aceuticals,Berlin,G

ermany

Sodiumiodide

Iodide−

1127

StudyI:134

Sigma-A

ldrich,St.Louis,MO

,USA

Om

niscanT

MG

adodiamide

0574

StudyI:180

Am

ershamH

ealth,Oslo,N

orway

Visipaque

TM

Iodixanol0

1550Study

III:21G

EH

ealthcareA

S,Oslo,N

orway

aIntra-articularinjection:2.5

mM

;Intravenousinjection:0.2

mm

ol/kg,500m

M



matrix for FEA of diffusion contained concentration values of eachdepth-wise pixel layer at different time points. Data analysis wasperformed using MATLAB (v. R2007b, MathWorks Inc., Natick, MA,USA).

5.2.2 X-ray microtomography

CECT measurements in studies II and III were conducted with amicroCT instrument (SkyScan-1172; SkyScan, Kontich, Belgium)with X-ray tube voltage of 100 kV and an isotropic voxel size of29.8 × 29.8 × 29.8 µm3. In study II, the samples were first imagedwithout contrast agent (ioxaglate, Table 5.3). After immersion intothe contrast agent, the samples were continuously imaged for 2.5h and then at the following time points: 3, 4, 5, 6, 7, 8, 9, 10, 11,12, 24, 26, 28, 30, 32, 34 and 36 h. In study III, the samples werefirst imaged when immersed in phosphate buffered saline (PBS)containing inhibitors of metalloproteinases. Then, after replacingPBS with contrast agent (ioxaglate and iodixanol, Table 5.3), thesamples were imaged after 5 min of immersion. Finally, the sampleswere imaged after 24 h of immersion.

In study II, a 1-mm-thick coronal slice at the middle of the lesionwas extracted for further analysis. The linear attenuation coefficientwas measured for two regions of interest (ROIs): one in the middleof the repaired cartilage and another in the adjacent intact cartilage.For the analysis of the subchondral bone, two cylindrical ROIs(diameter: 3 mm, height: 2 mm) were extracted below the repair andadjacent intact tissue. A global threshold technique was applied inthe segmentation of the bone [150]. The following parameters weredefined: bone volume fraction (bone volume (BV) / tissue volume(TV), %), bone surface to volume ratio (BS/BV, 1/mm) and bonemineral density (BMD, g/cm3). The analysis was conducted withCT Analyser software (v. 1.10.1.0; SkyScan).

In study III, the full cartilage thickness ROI was placed to themiddle of the sample (average of 100 coronal slices, width of 100pixels). A depth-wise X-ray absorption profile was achieved by



averaging the pixel rows of the ROI. The actual contrast agent dis-tribution after 24 h immersion time was determined by subtractingthe pre-contrast profile from the profile acquired at 24 h. The con-trast agent concentration in cartilage was normalized against thecontrast agent concentration in the immersion solution. Analyseswere conducted with custom MATLAB (version R2010a, MathWorksInc., Natick, MA, USA) scripts.

5.2.3 Delayed quantitative CT arthrography

In study IV, each knee was imaged with a clinical 64-slice CT scanner(DiscoveryTM PET/CT 690, GE Medical Systems, Waukesha, WI,USA) with X-ray tube voltage of 100 kV and an isotropic voxel sizeof 312 × 312 × 312 µm3. Before CT scan, the ioxaglate – Gd-DTPA2−

contrast agent mixture (Table 5.3) was injected intra-articularly. Gd-DTPA2− was included into the mixture as the patients were also MRimaged using the dGEMRIC protocol. Between the injection andimaging, the patient performed active flexion-extension exercises ofthe knee for 5 minutes to enable effective distribution of the contrastagent into joint surfaces. The CT scan at 5 minutes after injectioncorresponded a conventional CT arthrography. Another time pointat 45 minutes was selected based on a previous study [86] suggestingthat the maximum contrast agent concentration in cartilage can beachieved between 30 and 60 minutes after injection.

dQCTA analysis was conducted with Analyze software (Ana-lyze 10.0, AnalyzeDirect, Inc., Overland Park, KS, USA). Seven 3Dcartilage ROIs were extracted: medial condyles of tibia and femur(TMC and FMC), lateral condyles of tibia and femur (TLC and FLC),medial and lateral trochlear grooves (MTG and LTG) and patella(PAT). In addition, a ROI from the largest and clearest synovial fluid(SF) volume was selected. The mean X-ray attenuation values ofthe cartilage and SF ROIs at 5 and 45 minutes after contrast agentinjection were calculated. Contrast agent concentrations in SF wereused to normalize the cartilage parameters.



5.3 DELAYED GADOLINIUM-ENHANCED MRIOF CARTILAGE

In study IV, each knee was imaged three times using a 3T MRscanner (Siemens Skyra, Siemens Healthcare, Erlangen, Germany)(Table 5.2). In the first session, pre-contrast T1 mapping (single-slice,medial and lateral condyles) was performed before intravenous (IV)injection of 0.2 mmol/kg of Gd-DTPA2− (Table 5.3). Subsequently,patient performed active flexion-extension exercises of the knee for5 minutes and walked for 5 minutes. T1 mapping was repeated at90 minutes after the IV injection [43, 176].

Two weeks later, T1 mapping was performed after intra-articular(IA) injection of ioxaglate – Gd-DTPA2− contrast agent mixture(Table 5.3). Ioxaglate was included in the mixture, as the knees werealso CT imaged using the dQCTA protocol. Again, the post-contrastT1 mapping was performed at 90 minutes after the IA injection.

T1 maps were generated using an in-house application createdwith MATLAB (version 7.9.0; MathWorks inc., Natick, MA, USA).Six ROIs were extracted from manually segmented cartilage (MTG,LTG, TLC, FLC, TMC and FMC) and one ROI from SF. Patella wasnot properly visualized in the dGEMRIC images, and was excludedfrom the analyses. The mean T1 relaxation time (i.e. dGEMRIC-index) was calculated for each cartilage ROI. In addition, ∆R1 wascalculated for each ROI (equation 5.1).

∆R1 =

(1

T1,Gd− 1

T1,0

)(5.1)

5.4 ARTHROSCOPY

In study IV, routine arthroscopic examination of the knee was con-ducted for the patients after dQCTA and dGEMRIC. The cartilageappearance at predetermined sites of knee joint (TMC, FMC, TLC,FLC, MTG, LTG and PAT) was classified according to the Interna-tional Cartilage Repair Society (ICRS) grading system [28] by anexperienced orthopaedic surgeon (Table 5.4).



Table 5.4: ICRS classification.

ICRS grade Cartilage appearance0 Normal1 Nearly normal. Superficial lesions, soft indentation

and/or superficial fissures and cracks.2 Abnormal. Lesions extending down to <50% of

cartilage depth.3 Severely abnormal. Cartilage defects extending

down to >50% of cartilage depth as well as downto calcified layer and down to but not through thesubchondral bone. Blisters are included in thisgrade.

4 Severely abnormal.

5.5 SIMULATION OF CONTRAST AGENT DIFFUSION

In study I, the distribution of contrast agent in cartilage was simu-lated with a one-dimensional (1D) FE model (COMSOL 3.5, COM-SOL Inc., Burlington, MA, USA). The length of the 1D FE-geometrycorresponded to each cartilage disc thickness obtained from pQCTimages. The FE-geometry consisted of 15 1D-elements evenly dis-tributed along the cartilage depth. The COMSOL software solvesdiffusive mass transport from the Fick’s second law (equation 3.11).Diffusion was allowed only from one side of the sample (superfi-cial/deep cartilage), and the contrast agent concentration on thatdiffusion interface was assumed to be constant. To determine thediffusion coefficient (D), the model was fitted to the experimentallydetermined contrast agent concentration averaged over the full tissuedepth at each time point. The unconstrained nonlinear minimizationroutine of MATLAB (version R2007b; The MathWorks Inc., Natick,MA, USA) was used for optimization of the diffusion coefficient andthe concentration on the diffusion interface by minimizing the meansquare error between the experimental and simulated concentrationsat different time points.

In study II, D of contrast agent was determined as described



above, but based on the average values of the linear attenuationcoefficient at each time point.

5.6 REFERENCE METHODS

In studies II and III, the CECT parameters were compared with thereference histological (II and III), biochemical (III) and mechanical(II) parameters of cartilage. Histological analyses included polarizedlight microscopy (PLM) (II), digital densitometry (DD) (II and III)and Fourier transform infrared imaging (FTIRI) (III). The referencemethods will be described in the following sections.

5.6.1 Histological and biochemical analyses

In study II, samples were prepared after the CECT experiments forhistological analyses including PLM of collagen fibril orientationand DD of FCD distribution. In study III, the samples were cut intotwo halves. One half was prepared for the biochemical analysis ofwater content, collagen and cross-link concentrations, and the otherfor histological analysis of the FCD distribution with DD, and PGand collagen distributions with FTIRI.

Before the preparation of the microscopy sections, samples werefixed in 10% formalin for 48 hours, decalcified with 10% EDTAand 4% formaldehyde in 0.1 mol/L sodium phosphate buffer, andembedded in paraffin. 3- µm-thick sections were used in DD, 5-µm-thick sections in FTIRI, and 7- µm-thick sections in PLM.

Safranin O -stained sections were used to determine FCD distri-bution in cartilage samples by means of DD [82, 83]. A CCD-camera(CH250, Photometrics, Tucson, AZ) was used for the measurementsof the optical density of the Safranin O -stained sections.

PLM measurements of collagen fibril orientation angle and paral-lelism index (PI) maps were conducted for unstained sections usinga Leitz Ortholux BK II POL microscope (Leitz Wetzlar, Wetzlar, Ger-many) [144]. The orientation angle is defined as the angle betweenthe articular surface and the long axis of a fibril. PI describes the de-gree of structural anisotropy of collagen fibril network. The highest



PI is evidence of a structure where collagen fibrils are oriented inthe same direction, whereas the lowest PI value indicates randomlyorganized collagen fibrils.

Spatial distributions of the PG and collagen concentrations ofcartilage samples were determined with FTIRI using the PerkinElmer Spotlight 300 FTIRI system (PerkinElmer, Shelton, CO, USA)[34, 145]. The absorption spectra of the collagen and PG were at1585–1720 cm−1 and 984–1140 cm−1, respectively.

The water content of the samples was determined from the dif-ference between the wet weight and dry weight after freeze-drying.Subsequently, the samples were hydrolyzed with hydrochloric acid(HCl). After this and drying by evaporation and dissolving thesamples into water, the spectrophotometric assay for collagen spe-cific amino acid hydroxyproline was performed for the estimationof the collagen concentration [40]. High-performance liquid chro-matography was used to measure the concentrations of hydroxyly-sylpyridinoline (HP), lysylpyridinoline (LP) and pentosidine (Pent)cross-links [12, 89].

5.6.2 Mechanical testing

In study II, the samples were mechanically tested using a custom-made material testing device (Figure 5.1) which consisted of a loadcell (Sensotec, Columbus, OH, USA), a precision motion controller(PM500-C, Newport, Irvine, CA, USA) and a computerized dataacquisition system [187]. Mechanical measurements were carriedout using a stepwise (3 steps) creep indentation protocol [120]. Thefirst step was considered a pre-step to confirm full contact betweencartilage and impermeable, plane-ended indenter (radius: 272 µm).The stress was increased by 84 kPa in every step, and the deformationafter each step was recorded for 1200 seconds.

The experimental creep behaviour of each sample was analyzedusing the finite element method (FEM). The aim was to find outthe specific contributions by the GAGs, collagen network, and fluid.To achieve this, a fibril-reinforced poroelastic model [91, 103] was



Figure 5.1: A) Schematic illustration of the custom made material testing device. B) Increep measurement, the deformation is recorded while applying a constant load.

constructed using Abaqus (version 6.9 EF1; Dassault Systémes Simu-lia Corp., Providence, RI, USA). Articular cartilage was modeled asa biphasic material consisting of solid and fluid phases. The fluidfraction was adjusted to 85 % [91, 103]. The solid phase consistedof a non-fibrillar and fibrillar part corresponding to the mechan-ical effects of GAGs and the collagen network, respectively. Thenon-fibrillar matrix was assumed to be hyperelastic (neo-Hookean)characterized with Young’s modulus Enf and Poisson’s ratio ν. ThePoisson’s ratio of the non-fibrillar matrix was fixed to be 0.42 inall analyses [91, 103]. The behaviour of the collagen fibrils was ex-pressed with the fibril network modulus Ef, and the fluid flow wasdescribed with the permeability k.

Cartilage geometry was constructed with axisymmetric 4-nodecontinuum elements with pore pressure. Cartilage edge and surface(except for the cartilage-indenter contact) were set to be permeable(zero pore pressure). The bottom of the cartilage was fixed in x-and y-directions, and the movement at x-direction was preventedat the axis of symmetry. The indenter was modeled as a linearlyelastic material (E: 60 GPa, ν: 0.3), and the cartilage-indenter contactwas assumed to be frictionless. The step stress, implemented in



the model as a contact stress between cartilage and indenter, andcreep time were the same as in the experiments. The thicknessof cartilage layer of each sample was obtained from the microCTimages. The unconstrained nonlinear optimization routine of theMatlab Optimization Toolbox (version R2007b; The MathWorks, Inc.,Natick, MA) was used to fit the following parameters: En f , E f , andk. During fitting, the mean square error between the experimentaland simulated displacements was minimized.

5.7 STATISTICAL ANALYSES

The non-parametric Wilcoxon signed-ranks test was applied forall statistical comparisons in studies I, II and III. The differencesbetween the diffusion coefficients of different contrast agents anddifferent diffusion directions were compared in study I. In addition,false discovery rate control was used to adjust for multiple compar-isons. The differences between the parameter values determined forthe repair tissue and the adjacent intact tissue were compared instudy II. In study III, the differences between the properties of thereference and threose treated samples were compared. Spearman’scorrelation coefficients were determined between different parame-ters in studies II and III. Either Spearman’s or Pearson’s correlationcoefficient was determined in study IV, depending on the normalityof the distribution of parameter value tested using the Kolmogorov-Smirnov test. The statistical significances of differences betweenthe correlation coefficients were tested using the Fisher’s transform,and group comparisons were performed with the Kruskal-Wallistest in study IV. SPSS (version 14.0 SPSS Inc., Chicago, IL, USA),IBM SPSS Statistics (IBM SPSS Statistics 19.0.0.1, SPSS Inc., IBMCompany, Armonk, NY, USA), R (version 2.7.2 the R Foundationfor Statistical Computing, Vienna, Austria) and MATLAB (versionR2010a, MathWorks Inc.) were used in statistical analyses.


6 Results

The most important results of studies I–IV are summarized in thischapter. The more detailed results can be found in the originalpublications in the appendices.

6.1 CONTRAST AGENT DIFFUSION THROUGHSUPERFICIAL AND DEEP CARTILAGE

In study I, diffusion coefficients of four contrast agents in full-thickness cartilage were determined by analyzing CECT imagesby using the finite element method. Diffusion was significantly(p < 0.05) faster through articular surface than through deep car-tilage with all contrast agents except for iodide (Table 6.1). Thesimulated concentration values were in accordance with the experi-mental counterparts (Figure 6.1).

Table 6.1: Diffusion coefficients (mean ± SD) in articular cartilage for four contrast agentswith different mass and charge. Diffusion was significantly faster through articular surfacethan through deep cartilage with gadopentetate, gadodiamide and ioxaglate. Iodide diffusedsignificantly faster than ioxaglate and gadodiamide through the articular surface. Throughdeep cartilage, iodide diffused significantly faster than all the other contrast agents. (StudyI)

Diffusion coefficients (µm2/s)Contrast agent (mass, charge) Superficial to deep Deep to superficial

Iodide (127 g/mol, −1) 474.7 ± 251.7 292.3 ± 80.5Gadopentetate (548 g/mol, −2) 253.7 ± 224.6∗ 42.6 ± 45.2Gadodiamide (574 g/mol, 0) 161.1 ± 52.2∗ 53.9 ± 17.4Ioxaglate (1269 g/mol, −1) 142.8 ± 62.8∗ 41.8 ± 52.6∗p < 0.05 Wilcoxon signed-ranks test, n = 6 for each group.



Figure 6.1: Experimental (mean ± SD) and simulated normalized contrast agent concen-trations as a function of time. A) The contrast agent was allowed to penetrate through thearticular surface (n = 6 for each contrast agent). B) The contrast agent was allowed topenetrate through deep cartilage (n = 6 for each contrast agent). (Study I) In the originalpublication I the caption of Figure 2 mistakenly indicates that the contrast agent wouldhave been allowed to penetrate through the articular surface in six samples. In reality,the data in Figure 2 includes all the samples (n = 12 for each contrast agent) and bothdiffusion directions. (Study I)


Results

6.2 CECT OF SPONTANEOUSLY REPAIREDOSTEOCHONDRAL INJURIES

The diffusion coefficient of ioxaglate was significantly (p < 0.05)higher in spontaneously repaired equine cartilage than in the adja-cent intact tissue. However, contrast agent concentration in equilib-rium (i.e. X-ray attenuation) was not significantly different betweenrepaired and intact tissue. The structure and density of the sub-chondral bone under repaired and intact cartilage were significantly(p < 0.05) different. The moduli of the nonfibrillar matrix and thefibril network of the repair cartilage were significantly (p < 0.05)inferior to those determined for the adjacent intact tissue (Table6.2). Collagen fibril orientation and parallelism, and optical den-sity describing the GAG content of the tissue revealed also inferiorquality of the repair cartilage (Figure 6.2). The contrast agent dif-fusion coefficient correlated significantly with the modulus of thenonfibrillar matrix (R = −0.66, p < 0.01) and the parallelism index(R = −0.59, p < 0.05).



Table 6.2: Properties of spontaneously repaired cartilage and subchondral bone underrepair cartilage (mean ± 95% confidence interval) were significantly different from those ofadjacent intact tissue. (Study II)

Intact (n = 7) Repair (n = 7)D (µm2/s) 29.3 (7.59, 50.9) 107.9 (62.4, 153.3)∗

Enf (MPa) 0.65 (0.47, 0.83) 0.15 (0.08, 0.22)∗

Ef (MPa) 10.1 (8.08 12.0) 4.70 (2.21, 7.19)∗

αde(1 · 10−3/cm) 0.83 (0.74, 0.93) 0.75 (0.69, 0.82)αnc(1 · 10−3/cm) 0.69 (0.58, 0.80) 0.55 (0.47, 0.63)BV/TV (%) 87.2 (81.2, 93.1) 67.7 (55.8, 79.7)∗

BS/BV (1/mm) 5.12 (3.90, 6.34) 9.39 (7.04, 11.7)∗

BMD (g/cm3) 0.84 (0.81, 0.86) 0.72 (0.69, 0.74)∗

D=Diffusion coefficient, Enf=Modulus of the nonfibrillar matrix,Ef=Modulus of the fibril network, αde=Attenuation coefficient in dif-fusion equilibrium, αnc=Attenuation coefficient in native cartilage,BV/TV=Bone volume fraction (Bone volume/tissue volume), BS/BV=Bonesurface/volume ratio, BMD=Bone mineral density, ∗p < 0.05 Wilcoxonsigned-ranks test

6.3 EFFECT OF COLLAGEN CROSS-LINKING ONCONTRAST AGENT DIFFUSION

In study III, the concentrations of the pentosidine and lysylpyridi-noline cross-links were significantly (p < 0.05, n = 13) higher inthe threose-treated samples. The partition of anionic ioxaglate wassignificantly (p < 0.05, n = 7) lower in the threose-treated than inthe reference samples, while the threose-treatment did not affect thepartitioning of the non-ionic iodixanol (Figure 6.3). The FCD waselevated significantly (p < 0.05, n = 13) after the threose-treatmentin the middle and deep zones of cartilage. However, there was nosystematic difference in the GAG concentration of the reference andtreated samples.


Results

Figure 6.2: Orientation angle (A) and parallelism index (B) of collagen fibrils and opticaldensity (C) indicating spatial glycosaminoglycan content (mean ± 95% confidence interval).Significant statistical difference (p < 0.05, Wilcoxon signed-ranks test) between repair andadjacent intact tissue is marked with ∗. (Study II)



Figure 6.3: A) Partition of anionic ioxaglate in the threose-treated samples was significantlylower than that in the reference samples through the entire tissue depth. ∗p < 0.05, n = 7B) There was no cross-linking induced difference in partition of iodixanol (n = 6). (mean± 95% confidence interval) (Study III)

6.4 DQCTA AND DGEMRIC – ASSOCIATION TOARTHROSCOPIC FINDINGS

Delayed quantitative CT arthrography (dQCTA), intravenous andintra-articular dGEMRIC, and arthroscopic evaluation were per-formed for 11 patients in study IV. According to visual evaluation ofcartilage lesions, CT image acquired at 5 minutes after the contrastagent injection had the best diagnostic quality (Figure 6.4). Sig-nificant correlations were found between dGEMRIC and dQCTAparameters. ∆R1,IV (Figure 6.5) and ∆R1,IA/∆R1,SF correlated signifi-cantly with dQCTA parameter C45/SF45 (C=cartilage, SF=synovialfluid): (R = 0.72, p < 0.01) and (R = 0.70, p < 0.01), respectively.

dQCTA and dGEMRIC parameters were not significantly differ-ent between groups with different ICRS grade. However, C45/SF45

revealed a positive trend in conjuction with increasing ICRS grade(Figure 6.6).

An association between IA and IV dGEMRIC was found only af-ter normalizing ∆R1,IA with ∆R1,SF: R = 0.52 between ∆R1,IA/∆R1,SF

and ∆R1,IV (p < 0.01).


Results

Figure 6.4: MR (A,C,E) and CT (B,D,F) images showing cartilage lesion with ICRS grade2 (arrow). A) Anatomical MR image B) CT at 5 min after contrast agent injection, C) Pre-contrast MR image, D) CT at 45 min after contrast agent injection, E) T1 relaxation timemap overlaid on top of a T1-weighted MR image 90 min after IV contrast agent injection(T1,Gd) F) dQCTA map of normalized X-ray attenuation in cartilage at 45 minutes aftercontrast agent injection (C45/SF45) overlaid on top of an anatomical CT image. (Contrastof the images has been adjusted.) (Study IV)



Figure 6.5: Significant correlation between normalized mean X-ray attenuation in cartilageat 45 min after contrast agent injection (C45/SF45) and ∆R1,IV. Linear fit is for illustrativepurposes. The distribution of the ICRS grades was: grade 0; n=19, grade 1; n=19, grade 2;n=19, grade 3; n=2. (Study IV)

Figure 6.6: Normalized dQCTA parameter C45/SF45 in different ICRS grade groups. Inspite of a positive trend in conjuction with increasing ICRS grade there were no significantdifferences between the groups. Grades 2 and 3 were merged together due to the low numberof grade 3 lesions. (Study IV)


7 Discussion

In this thesis, the diffusion of X-ray contrast agents was investi-gated in intact, spontaneously healed and artificially cross-linkedcartilage in vitro. Furthermore, the clinical potential of delayed CTarthrography was evaluated in vivo. In study I, the diffusion of stud-ied contrast agents was found to occur faster through superficialcartilage than through deep cartilage. In study II, CECT enabledquantitative distinguishing of spontaneously healed cartilage andsubchondral bone from adjacent intact tissues in the equine inter-carpal joint. In study III, the effects of increased cross-linking ofcollagen on diffusion of anionic and non-ionic contrast agents werefound to be more related to the changes in tissue FCD than changesin steric hindrance. In study IV, dGEMRIC and dQCTA (i.e. CECT invivo) were found to agree but not to be in line with the arthroscopicfindings.

7.1 CONTRAST AGENT DIFFUSION THROUGHSUPERFICIAL CARTILAGE IS FASTER THANTHROUGH DEEP CARTILAGE

In study I, the diffusion coefficients in cartilage were determinedfor four contrast agents. Diffusion was allowed either throughthe superficial cartilage or through deep cartilage. The smallestmolecule, iodide, displayed the highest diffusion coefficient whilethe largest molecule, ioxaglate, showed the lowest diffusion coef-ficient. This is in line with previous studies that have reportedsmaller diffusion coefficients with an increasing size of the diffusingmolecule [3, 100, 101, 107, 179–182]. Two molecules with equivalentsizes, i.e. gadopentetate and gadodiamide, exhibited no significant(p > 0.05) differences in their diffusion coefficients even though themolecules possess different charges.

Diffusion coefficients of gadopentate, gadodiamide and ioxaglate



for penetration through superficial cartilage were higher than thosemeasured in penetration through deep cartilage. This result was ex-pected and can be explained by the inhomogeneous composition ofcartilage: the higher water content and the lower GAG concentrationin the superficial cartilage. This is supported by previous studiesthat have reported higher diffusion coefficients with increasing wa-ter content and decreasing GAG concentration [49, 106–108, 181].Differences in collagen network structure and orientation have beenproposed to explain the changes in diffusion coefficient of largedextran molecules (40 kDa – 500 kDa) along the tissue depth whilethe GAG concentration has been postulated to mainly hinder dif-fusion of smaller dextran molecules (3 kDa) [100]. In contrast, aCECT study examining the diffusion of ioxaglate (≈1.3 kDa) andgadopentetate (≈0.5 kDa) reported that it took over 8 h to reachthe equilibrium in intact and proteoglycan depleted cartilage [79].This suggests that collagen may also hinder the diffusion of smallermolecules.

7.2 CECT ENABLES DETECTION OF REPAIREDOSTEOCHONDRAL INJURIES

Spontaneously healed cartilage tissue could be quantitatively dis-tinguished from the adjacent intact tissue in study II. The signifi-cantly higher contrast agent diffusion coefficient in repaired tissuemay be attributable to the lower GAG content in the repair tis-sue [49, 106–108, 181]. In addition, the abnormal structure of thecollagen network in repaired tissue may have contributed to thehigher diffusion coefficient values. The collagen fibrils in repairedcartilage were not oriented tangentially in the superficial tissuewhich can partially explain the present findings, since diffusion inparallel to the collagen fibrils is faster than diffusion in a perpendic-ular direction to the fibrils [101].

The concentration of the contrast agent at diffusion equilibriumwas not significantly different between repaired and intact cartilagedespite the lower GAG content in the repaired tissue. This suprising


Discussion

result may be due to the abnormal organization of the collagennetwork in repaired tissue which could hinder the mobility of thecontrast agent. This is supported by a recent contrast-enhanced MRIstudy where the alterations in the collagen network were reported tohave an effect on the accumulation of contrast agent in cartilage [200].

The modulus of the nonfibrillar matrix in the repaired tissuewas smaller than in the intact tissue. The GAG content is known tobe associated to the modulus of the nonfibrillar matrix [91]. Thus,the lower GAG content detected in the repaired tissue supportsthis result. The abnormal organization of the collagen network inthe repaired tissue may be responsible for the smaller modulus ofthe fibrillar matrix. In particular, the lack of the superficial tangen-tially oriented fibril layer may contribute to the inferior mechanicalproperties of the repaired tissue [61, 92, 156].

The new mechanical environment of the subchondral bone un-derneath the repaired cartilage affects the healing and remodelingresponse of bone, and this may be responsible for the lower BMDin that area [59]. The reduced values of BV/TV and BS/BV seen insubchondral bone under repaired cartilage are typical signs of thesubchondral reaction related to an osteochondral injury [74,191,192].

7.3 CROSS-LINKING RELATED CHANGES IN FCD HAVEGREATER EFFECT ON THE CONTRAST AGENTDIFFUSION THAN CHANGES IN STERIC HINDRANCE

In study III, diffusion of ionic and non-ionic contrast agents wasinvestigated in cartilage treated with threose to increase cross-linkingof collagen. As expected in the light of the results of our earlierstudy [87], partition of anionic ioxaglate was significantly lower inthe threose-treated tissue than in the reference samples. However,there did not seem to be any difference in the partition of non-ionic iodixanol. This suggests that the cross-links do not cause asignificant steric hindrance for these contrast agent molecules.

The elevated negative FCD in the threose-treated samples sup-ports the finding of decreased partitioning of ioxaglate. FCD is



traditionally assumed to be linearly related to the GAG concentra-tion of the tissue [60, 109, 193]. However, the GAG concentrationdid not exhibit a systematic change following the threose-treatment.Pentosidine cross-links, which were significantly increased afterthreose-treatment, are formed between arginine and lysine residuesin collagen [155]. This leads to the loss of positive charges of theseresidues, causing the total negative FCD in cartilage to increase [62].The cross-linking related increase in FCD challenges the traditionalinterpretation of the results of dGEMRIC and CECT, since despitethe increased FCD, the GAG concentration did not change accordingto these present findings and our previous study [87]. Since theconcentration of the pentosidine cross-links is reported to increasewith ageing [11], the age of a patient can affect the results achievedby imaging techniques based on the use of charged contrast agents.

7.4 DQCTA IS IN AGREEMENT WITH DGEMRIC IN VIVO

The results obtained with dGEMRIC (intravenous and intra-articular)and dQCTA were found to be in agreement (study IV). This wasexpected, although the contrast agents (gadopentetate and ioxaglate)differ in terms of charge and molecular mass (Table 5.3) and differ-ences in their penetration into cartilage have been reported [15, 165].However, in order to achieve agreement between the techniques,dQCTA parameter and ∆R1,IA had to be normalized with the con-trast agent concentration in synovial fluid and ∆R1,SF, respectively.Hence, based on the present results, it is important to consider thecontrast agent concentration in synovial fluid after intra-articularinjection of contrast agent.

The results from dQCTA and dGEMRIC were not related toarthroscopic ICRS grading. This is in line with two previous dGEM-RIC studies in which the dGEMRIC-index did not correlate witharthroscopic grading [132, 134]. This finding may be partially ex-plained by the fact that while the remaining cartilage is analyzed indGEMRIC and dQCTA, it is the absence of cartilage which definesthe ICRS grade.


Discussion

dGEMRICIV and dGEMRICIA correlated moderately when the∆R1,IA was normalized with ∆R1,SF. The other examined dGEMRICparameters showed no significant correlations. This finding may beexplained by differences in IV and IA administration techniques.First, the volume of the contrast agent is different in IV and IAinjections. With an IV injection, it is the weight of the patient whichdetermines the final volume to be injected whereas with an IAinjection, the volume is the same for each patient. Second, afterintra-articular injection, the contrast agent is diluted in synovialfluid, the volume of which is different between patients. Third, thereare differences in the pharmacokinetics of contrast agent betweenIA and IV administrations, although diffusion from subchondralbone after IV injection was found to be negligible in a recent in vivostudy [66]. This is also supported by in vitro studies [152, 165].

7.5 LIMITATIONS

There are some limitations in studies I–IV that should be addressed.The rather low number of samples in the studies was accepted forpractical reasons. However, it was ensured that in the in vitro studiesthe control and experimental samples were paired, and the testgroups included independent sample pairs.

The main limitation of study I was the rather low number ofimaging time points, partly because of long imaging time requiredto achieve high resolution with an acceptable signal-to-noise ratio.A higher number of time points, especially in the early phase ofdiffusion, could have improved the accuracy of the FE model fitting.In our MRI study of gadopentetate diffusion in bovine cartilage,however, imaging was conducted every 6 minutes for 18 hours [152].The diffusion data gathered in that study was analyzed in two ways:using all the data points in the fitting and using only those datapoints that corresponded to the time points of study I (1, 5, 9, 16,18 h). A good linear correlation was found between the diffusioncoefficients analyzed in these two ways (R = 0.94). Thus, the numberof imaging time points in study I could be considered as adequate.



In study II, the mechanical properties of repair cartilage werefound to be inferior when compared to those of intact cartilage.When testing the samples with creep indentation, it was challengingto achieve good contact between cartilage and the indenter. There-fore, the first compressive step was considered a prestep in thesimulations of creep behaviour of both repair and intact cartilage.Another limitation of study II was that the sample-specific materialproperties (depth-dependent collagen orientation, collagen concen-tration and PG concentration) of cartilage were not implementedin the FE model. Implementation of the sample-specific propertiescould have affected the simulated results, but would have made themodeling more complex.

In study III, the effects of age-related changes in the collagennetwork on the diffusion of the contrast agent were investigated. Thecross-links were created chemically using threose but these may notperfectly mimic the changes in collagen network occurring duringageing. Further, cartilage undergoes also other changes with age, e.g.surface fibrillation, decreased size of PG aggregates, and decreasedwater content [24,111]. However, incubation in threose did representa feasible way to increase pentosidine cross-links in order to modifythe structure of the collagen network in a way typical of naturalageing [11].

Due to the complicated experimental set-up carried out withclinical patients in study IV, certain limitations had to be accepted.First, the time delay between contrast agent injection and imagingwas different in dGEMRIC and dQCTA for practical reasons. Theselection of the time points was based on the literature [43, 86, 176],and the contrast agents for dGEMRICIA and dQCTA were chosento be injected at the same time for ethical reasons. Second, in thefirst patient, the IA injection was given to the suprapatellar recess,which is the most common site for intra-articular injections [54, 78].Unfortunately, this resulted in an uneven distribution of contrastagent in the joint cavity. For the remainder of the patients, the con-trast agent was injected directly into the joint space between femurand tibia. Third, although there were differences in the segment-


Discussion

ing procedures of dGEMRIC and dQCTA images, the agreementcould be confirmed visually. It was possible to use semi-automatedsegmentation in dQCTA analysis due to higher contrast betweencartilage and adjacent tissues. In addition, two-dimensional (2D)ROI was segmented in dGEMRIC in contrast to 3D ROI in dQCTAas acquisition of only single-slice in dGEMRIC enabled reasonableimaging times.

7.6 DQCTA – CURRENT STATUS

dQCTA is already a clinically available imaging technique. Asa result of the advances in CT technology, clinical microCT witha 41 µm isotropic voxel size which allows imaging of peripheraljoints is already on the market (XtremeCT, SCANCO Medical AG,Brüttisellen, Switzerland) [154]. Such a high level of resolution willalso allow simultaneous, reliable, and quantitative evaluation of thebone microstructure [73].

HexabrixTM (ioxaglate) is a widely used X-ray contrast agentand it is appropriate for dQCTA. In study IV, the concentration ofHexabrixTM was 105 mM (diluted to 25% of full strength HexabrixTM).However, full strength HexabrixTM injections have been applied andthese would enable better delineation of cartilage in the arthro-graphic images. Full strength HexabrixTM with osmolality of 600mOsm/kg [46] is hyperosmolalic when compared to the synovialfluid with osmolality of approximately 300-400 mOsm/kg [19, 63].Hyperosmolalic environment causes softening of the cartilage dueto decreased osmotic pressure [48,90]. Softening of the cartilage mayincrease the risk of mechanical damage and increased cell death ifcartilage is subjected to loading after injection of hyperosmolariccontrast agent [186]. This is important to be borne in mind sinceusually the joint needs to be exercised after injection to ensure aneven distribution of contrast agent. However, in clinical practiceintra-articular injection of full strength HexabrixTM may be consid-ered safe if the joint is not heavily loaded and as contrast agentdilutes significantly when mixing with synovial fluid.



Since each patient has a different amount of synovial fluid, thecontrast agent concentration in the joint capsule varies among pa-tients. This should not affect the interpretation of dQCTA imagessince contrast agent concentration has been reported not to affect thepartitioning in cartilage in vitro [162]. However, in order to ensureadequate contrast, a high enough concentration and volume needs tobe injected. This is to assure that the contrast agent is not excessivelydiluted in synovial fluid. A volume of 20 ml was used in studyIV (concentration: 105 mM) and in an earlier study (concentration:210 mM) [86]. In those cases where the knee is very swollen, thesynovial fluid may need to be aspirated before injection of contrastagent [54].

Cationic contrast agents have shown potential for more sensitivedetection of changes in cartilage FCD. Cationic contrast agents havebeen reported to penetrate to higher concentrations in cartilage thananionic contrast agents which eases the segmentation of cartilage[14, 15, 75, 170]. Despite the positive charge strongly attracted bycartilage, the contrast agent has been reported to be washed out ofthe rabbit joint in vivo and to be safe according to histopathologicexamination of cartilage after imaging experiments [170].

It has been suggested that three acquisitions should be includedin the dQCTA protocol: non-contrast, arthrographic and delayed [86].In study IV, only arthrographic and delayed images were acquiredto minimize the radiation dose. According to the radiologist, thearthrographic image had the best diagnostic quality for detectionof cartilage lesions. However, in order to be able to evaluate thecondition of the cartilage surrounding the lesion, it was found nec-essary to use the delayed image. Due to rather slow diffusion ofcontrast agent into cartilage, only a very small amount of contrastagent can be expected to be present in cartilage at the moment ofacquisition of the arthrographic image. Thus, the necessity of thenon-contrast image can be questioned. The subtraction image isachieved by subtracting the arthrographic image from the delayedimage, and then the amount of contrast agent diffused into cartilagebetween these two time points (5 min and 45 min in study IV) can


Discussion

be calculated. This may serve as an adequate indicator of the stateof the cartilage.

The effective radiation dose received by a patient during onescan with the XtremeCT scanner mentioned earlier is less than 5 µSvbased on the information given by the manufacturer [154]. Thus, thetotal dose of two sequential scans would correspond approximatelyone day’s exposure to the background radiation in Finland.

7.7 DQCTA – FUTURE

dQCTA is a promising clinical method for quantitative diagnosis ofcartilage injuries and degeneration. However, further studies withlarger populations including healthy participants and patients withOA are needed to optimize the technique and to fully characterize itspotential. For example, a study with patients referred to total kneereplacements would allow biochemical and histological evaluationof cartilage as a reference to the results of dQCTA.

From the modeling perspective, development of a sophisticateddiffusion model would be important. With the aid of the model, itwould be possible to determine the individual contribution of dif-ferent constituents of cartilage to the diffusion process. The modelshould take into account the contribution of electrostatic interac-tions of contrast agent molecules with proteoglycans. Structuralorientation of the collagen network and the size of the diffusingmolecule should also be included into the model. Then, combina-tion of dQCTA and modeling would enable compositional imagingwhich would help in the early diagnosis of OA and evaluation ofcartilage repair.


8 Summary and conclusions

In this thesis, factors affecting diffusion and distribution of X-raycontrast agents in articular cartilage were investigated. Finally, CECTwas applied in vivo together with dGEMRIC, and their associationwas compared to the arthroscopic evaluation of the knee. The mainfindings of this thesis are summarized as follows:

1. The tissue-averaged axial diffusion coefficients of four contrastagents in cartilage could be determined. Diffusion throughthe articular surface was faster than through deep cartilage.This appeared to reflect the differences in composition alongthe cartilage depth: higher water content and lower GAGconcentration in superficial than in deep cartilage.

2. MicroCECT was able to quantitatively distinguish the sponta-neously healed repair site from adjacent intact tissue.

3. Cross-linking induced changes in cartilage fixed charge densityhad a greater effect on diffusion of anionic contrast agentcompared to the changes in steric hindrance.

4. dQCTA normalized with the contrast agent concentration insynovial fluid achieved the best agreement with dGEMRIC (in-travenous and intra-articular) at 45 minutes after intra-articularinjection of X-ray contrast agent. However, neither of the tech-niques revealed any correlation to arthroscopic findings. Thisis probably because the remaining cartilage is analyzed indQCTA and dGEMRIC, while it is the absence of cartilagewhich defines the ICRS grade.

5. Since normalizing with the contrast agent concentration insynovial fluid was found to be of great importance, it is pro-posed to be taken into account in future clinical application ofdQCTA.



dQCTA is a promising clinical method for quantitative diagnosisof cartilage injuries and degeneration. The further development ofdQCTA would require clinical studies with healthy and OA patients,and patients referred to total knee replacement which would per-mit the collection of tissue samples for reference measurements. Inaddition, the development of a composition-based diffusion model,which would take into account the individual contribution of differ-ent constituents of cartilage to the contrast agent diffusion, wouldenable compositional imaging when combined with dQCTA. Thiswould help in the early diagnosis of OA and in the evaluation ofcartilage repair.


References

[1] A. M. Aisen, W. J. McCune, A. MacGuire, P. L. Carson, T. M. Silver,S. Z. Jafri, and W. Martel, “Sonographic evaluation of the cartilageof the knee,” Radiology 153, 781–4 (1984).

[2] R. G. Allen, D. Burstein, and M. L. Gray, “Monitoring gly-cosaminoglycan replenishment in cartilage explants with gadolinium-enhanced magnetic resonance imaging,” J Orthop Res 17, 430–6(1999).

[3] R. V. Allhands, P. A. Torzilli, and F. A. Kallfelz, “Measurement ofdiffusion of uncharged molecules in articular cartilage,” Cornell Vet74, 111–23 (1984).

[4] R. Altman, E. Asch, D. Bloch, G. Bole, D. Borenstein, K. Brandt,W. Christy, T. D. Cooke, R. Greenwald, M. Hochberg, and et al.,“Development of criteria for the classification and reporting of os-teoarthritis. Classification of osteoarthritis of the knee. Diagnosticand Therapeutic Criteria Committee of the American RheumatismAssociation,” Arthritis Rheum 29, 1039–49 (1986).

[5] J. S. Apple, S. Martinez, M. B. Khoury, R. Royster, and S. Allen, “Acomparison of Hexabrix and Renografin-60 in knee arthrography,”AJR Am J Roentgenol 145, 139–42 (1985).

[6] J. P. Arokoski, M. M. Hyttinen, T. Lapveteläinen, P. Takacs, B. Kosz-taczky, L. Modis, V. Kovanen, and H. Helminen, “Decreased bire-fringence of the superficial zone collagen network in the canine knee(stifle) articular cartilage after long distance running training, de-tected by quantitative polarised light microscopy,” Ann Rheum Dis55, 253–64. (1996).

[7] G. A. Ateshian, “The role of interstitial fluid pressurization inarticular cartilage lubrication,” J Biomech 42, 1163–1176 (2009).

[8] G. A. Ateshian, L. J. Soslowsky, and V. C. Mow, “Quantitation ofarticular surface topography and cartilage thickness in knee jointsusing stereophotogrammetry,” J Biomech 24, 761–776 (1991).

[9] A. S. Aula, J. S. Jurvelin, and J. Töyräs, “Simultaneous computed to-mography of articular cartilage and subchondral bone,” OsteoarthritisCartilage 17, 1583–8 (2009).



[10] M. Baldassarri, J. S. Goodwin, M. L. Farley, B. E. Bierbaum, S. R.Goldring, M. B. Goldring, D. Burstein, and M. L. Gray, “Relationshipbetween cartilage stiffness and dGEMRIC index: correlation andprediction,” J Orthop Res 25, 904–12 (2007).

[11] R. A. Bank, M. T. Bayliss, F. P. Lafeber, A. Maroudas, and J. M. Tekop-pele, “Ageing and zonal variation in post-translational modificationof collagen in normal human articular cartilage. The age-relatedincrease in non-enzymatic glycation affects biomechanical propertiesof cartilage,” Biochem J 330 (Pt 1), 345–51 (1998).

[12] R. A. Bank, B. Beekman, N. Verzijl, J. A. de Roos, A. N. Sakkee, andJ. M. TeKoppele, “Sensitive fluorimetric quantitation of pyridiniumand pentosidine crosslinks in biological samples in a single high-performance liquid chromatographic run,” J Chromatogr B Biomed SciAppl 703, 37–44 (1997).

[13] P. N. Bansal, N. S. Joshi, V. Entezari, M. W. Grinstaff, and B. D.Snyder, “Contrast enhanced computed tomography can predict theglycosaminoglycan content and biomechanical properties of articularcartilage,” Osteoarthritis Cartilage 18, 184–91 (2010).

[14] P. N. Bansal, N. S. Joshi, V. Entezari, B. C. Malone, R. C. Stewart, B. D.Snyder, and M. W. Grinstaff, “Cationic contrast agents improve quan-tification of glycosaminoglycan (GAG) content by contrast enhancedCT imaging of cartilage,” J Orthop Res 29, 704–9 (2011).

[15] P. N. Bansal, R. C. Stewart, V. Entezari, B. D. Snyder, and M. W.Grinstaff, “Contrast agent electrostatic attraction rather than repul-sion to glycosaminoglycans affords a greater contrast uptake ratioand improved quantitative CT imaging in cartilage,” OsteoarthritisCartilage 19, 970–6 (2011).

[16] A. Bashir, M. L. Gray, R. D. Boutin, and D. Burstein, “Glycosamino-glycan in articular cartilage: in vivo assessment with delayedGd(DTPA)(2-)-enhanced MR imaging,” Radiology 205, 551–8 (1997).

[17] A. Bashir, M. L. Gray, and D. Burstein, “Gd-DTPA2- as a measure ofcartilage degradation,” Magn Reson Med 36, 665–73 (1996).

[18] A. Bashir, M. L. Gray, J. Hartke, and D. Burstein, “Nondestructiveimaging of human cartilage glycosaminoglycan concentration byMRI,” Magn Reson Med 41, 857–65 (1999).


References

[19] M. Baumgarten, R. D. Bloebaum, S. D. Ross, P. Campbell, andA. Sarmiento, “Normal human synovial fluid: osmolality andexercise-induced changes,” J Bone Joint Surg Am 67, 1336–9 (1985).

[20] A. Bedi, B. T. Feeley, and r. Williams, R. J., “Management of articularcartilage defects of the knee,” J Bone Joint Surg Am 92, 994–1009(2010).

[21] A. Benninghoff, “Form und Bau der Gelenkknorpel in ihrenBeziehungen zur Function. Erste Mitteilung: die modellierendenund formerhaltenden Faktoren des Knorpelreliefs.,” Z Anat 76, 43–63 (1925).

[22] M. A. Biot, “General theory of three-dimensional consolidation,”Journal of Applied Physics 12, 155–164 (1941).

[23] D. Boyd, “KNEE-JOINT VISUALIZATION A RoentgenographicStudy with Iopax,” J Bone Joint Surg Am 16, 671–80 (1934).

[24] K. D. Brandt and R. S. Fife, “Ageing in relation to the pathogenesisof osteoarthritis,” Clin Rheum Dis 12, 117–30 (1986).

[25] K. D. Brandt and E. Radin, “The physiology of articular stress:osteoarthrosis,” Hosp Pract (Off Ed) 22, 103–7, 110–2, 117–9 passim(1987).

[26] M. Brittberg, “Autologous chondrocyte implantation–technique andlong-term follow-up,” Injury 39 Suppl 1, S40–9 (2008).

[27] M. Brittberg, A. Lindahl, A. Nilsson, C. Ohlsson, O. Isaksson, andL. Peterson, “Treatment of deep cartilage defects in the knee withautologous chondrocyte transplantation,” N Engl J Med 331, 889–95(1994).

[28] M. Brittberg and C. S. Winalski, “Evaluation of cartilage injuries andrepair,” J Bone Joint Surg Am 85-A Suppl 2, 58–69 (2003).

[29] J. A. Buckwalter and H. J. Mankin, “Articular cartilage: Part I. Tissuedesign and chondrocyte-matrix interactions,” J Bone Joint Surg Am79-A, 600–611 (1997).

[30] J. A. Buckwalter and H. J. Mankin, “Articular cartilage: degenerationand osteoarthritis, repair, regeneration, and transplantation,” InstrCourse Lect 47, 487–504 (1998).

[31] J. A. Buckwalter and J. Martin, “Degenerative joint disease,” ClinSymp 47, 1–32 (1995).



[32] J. A. Buckwalter and J. A. Martin, “Osteoarthritis,” Adv Drug DelivRev 58, 150–67 (2006).

[33] J. A. Buckwalter, L. C. Rosenberg, and E. B. Hunziker, “Articularcartilage: Composition, structure, response to injury, and methods offacilitating repair,” in Articular cartilage and knee joint function: Basicscience and arthroscopy. (Raven Press Ltd., New York, 1990), pp. 19-56.

[34] N. P. Camacho, P. West, P. A. Torzilli, and R. Mendelsohn, “FTIR mi-croscopic imaging of collagen and proteoglycan in bovine cartilage,”Biopolymers 62, 1–8 (2001).

[35] W. Chan, P. Lang, M. Stevens, K. Sack, S. Majumdar, D. Stoller,C. Basch, and G. H.K., “Osteoarthritis of the knee: comparison ofradiography, CT, MR imaging to assess extent and severity,” AJR AmJ Roentgenol 157, 799–806 (1991).

[36] C. R. Chu, D. Lin, J. L. Geisler, C. T. Chu, F. H. Fu, and Y. Pan,“Arthroscopic microscopy of articular cartilage using optical coher-ence tomography,” Am J Sports Med 32, 699–709 (2004).

[37] M. D. Cockman, C. A. Blanton, P. A. Chmielewski, L. Dong, T. E.Dufresne, E. B. Hookfin, M. J. Karb, S. Liu, and K. R. Wehmeyer,“Quantitative imaging of proteoglycan in cartilage using a gadolin-ium probe and microCT,” Osteoarthritis Cartilage 14, 210–4 (2006).

[38] B. Cohen, W. M. Lai, and V. C. Mow, “A transversely isotropicbiphasic model for unconfined compression of growth plate andchondroepiphysis,” J Biomech Eng 120, 491–6 (1998).

[39] J. Crank, The mathematics of diffusion, 2nd ed. (Clarendon Press,Oxford, 1975).

[40] L. B. Creemers, D. C. Jansen, A. van Veen-Reurings, T. van denBos, and V. Everts, “Microassay for the assessment of low levels ofhydroxyproline,” Biotechniques 22, 656–8 (1997).

[41] M. D. Crema, F. W. Roemer, M. D. Marra, D. Burstein, G. E. Gold,F. Eckstein, T. Baum, T. J. Mosher, J. A. Carrino, and A. Guermazi,“Articular cartilage in the knee: current MR imaging techniques andapplications in clinical practice and research,” Radiographics 31, 37–61(2011).

[42] E. L. Cussler, Diffusion: mass transfer in fluid systems, 3rd ed. (Cam-bridge University Press, Cambridge; New York, 2009).


References

[43] L. Dahlberg, E. Lammentausta, C. Tiderius, and M. T. Nieminen,“In Vivo Monitoring of Joint Cartilage – Lessons to Be Learned byDelayed Gadolinium Enhanced Magnetic Resonance Imaging ofCartilage,” European Musculoskeletal Review 7, 58–62 (2012).

[44] M. De Filippo, A. Bertellini, F. Pogliacomi, N. Sverzellati, D. Corradi,G. Garlaschi, and M. Zompatori, “Multidetector computed tomogra-phy arthrography of the knee: diagnostic accuracy and indications,”Eur J Radiol 70, 342–51 (2009).

[45] F. G. Donnan, “The theory of membrane equilibria,” Chem Rev 1,73–90 (1924).

[46] Drugs.com, “Hexabrix Official FDA information, side effects anduses.,” http://www.drugs.com/pro/hexabrix.html (2010).

[47] Duodecim, Suomalainen Lääkäriseura, “Käypä hoito -suositus,Polvi- ja lonkkanivelrikko,” http://www.kaypahoito.fi/web/kh/suositukset/naytaartikkeli/./hoi50054 (2012).

[48] S. R. Eisenberg and A. J. Grodzinsky, “Swelling of articular cartilageand other connective tissues: electromechanochemical forces,” JOrthop Res 3, 148–59 (1985).

[49] R. C. Evans and T. M. Quinn, “Solute diffusivity correlates withmechanical properties and matrix density of compressed articularcartilage,” Arch Biochem Biophys 442, 1–10 (2005).

[50] D. Eyre, “Collagen of articular cartilage,” Arthritis Res 4, 30–5 (2002).

[51] D. T. Felson, “Does excess weight cause osteoarthritis and, if so,why?,” Ann Rheum Dis 55, 668–70 (1996).

[52] D. T. Felson, “Preventing knee and hip osteoarthritis,” Bull RheumDis 47, 1–4 (1998).

[53] D. T. Felson and T. Neogi, “Osteoarthritis: is it a disease of cartilageor of bone?,” Arthritis Rheum 50, 341–4 (2004).

[54] R. H. Freiberger and H. Pavlov, “Knee arthrography,” Radiology 166,489–92 (1988).

[55] J. A. Gagliardi, E. M. Chung, V. P. Chandnani, K. L. Kesling, K. P.Christensen, R. N. Null, M. G. Radvany, and M. F. Hansen, “Detec-tion and staging of chondromalacia patellae: relative efficacies ofconventional MR imaging, MR arthrography, and CT arthrography,”AJR Am J Roentgenol 163, 629–36 (1994).



[56] A. Gillis, A. Bashir, B. McKeon, A. Scheller, M. L. Gray, andD. Burstein, “Magnetic resonance imaging of relative glycosamino-glycan distribution in patients with autologous chondrocyte trans-plants,” Invest Radiol 36, 743–8 (2001).

[57] A. Gillis, M. Gray, and D. Burstein, “Relaxivity and diffusion ofgadolinium agents in cartilage,” Magn Reson Med 48, 1068–71 (2002).

[58] R. Glaser, Biophysics, Rev. 5th ed. (Springer, Berlin; New York, 2001).

[59] M. B. Goldring and S. R. Goldring, “Articular cartilage and subchon-dral bone in the pathogenesis of osteoarthritis,” Ann N Y Acad Sci1192, 230–7 (2010).

[60] G. Grushko, R. Schneiderman, and A. Maroudas, “Some biochemicaland biophysical parameters for the study of the pathogenesis ofosteoarthritis: a comparison between the processes of ageing anddegeneration in human hip cartilage,” Connect Tissue Res 19, 149–76(1989).

[61] F. Guilak, A. Ratcliffe, N. Lane, M. P. Rosenwasser, and V. C. Mow,“Mechanical and biochemical changes in the superficial zone of artic-ular cartilage in canine experimental osteoarthritis,” J Orthop Res 12,474–84. (1994).

[62] J. C. Hadley, K. M. Meek, and N. S. Malik, “Glycation changes thecharge distribution of type I collagen fibrils,” Glycoconj J 15, 835–40(1998).

[63] A. C. Hall, E. R. Horwitz, and R. J. Wilkins, “The cellular physiologyof articular cartilage,” Exp Physiol 81, 535–45 (1996).

[64] F. M. Hall and G. Wyshak, “Thickness of articular cartilage in thenormal knee,” J Bone Joint Surg Am 62, 408–13 (1980).

[65] L. Hangody and P. Fules, “Autologous osteochondral mosaicplastyfor the treatment of full-thickness defects of weight-bearing joints:ten years of experimental and clinical experience,” J Bone Joint SurgAm 85-A Suppl 2, 25–32 (2003).

[66] Z. K. Hawezi, E. Lammentausta, J. Svensson, L. E. Dahlberg, and C. J.Tiderius, “In vivo transport of Gd-DTPA(2-) in human knee cartilageassessed by depth-wise dGEMRIC analysis,” J Magn Reson Imaging34, 1352–8 (2011).


References

[67] C. W. Hayes, D. A. Jamadar, G. W. Welch, M. L. Jannausch, L. L.Lachance, D. C. Capul, and M. R. Sowers, “Osteoarthritis of theknee: comparison of MR imaging findings with radiographic severitymeasurements and pain in middle-aged women,” Radiology 237, 998–1007 (2005).

[68] M. Heliövaara, P. Slätis, and P. Paavolainen, “Nivelrikon esiintyvyysja kustannukset,” Duodecim 124, 1869–1874 (2008).

[69] Y. P. Huang and Y. P. Zheng, “Intravascular Ultrasound (IVUS): APotential Arthroscopic Tool for Quantitative Assessment of ArticularCartilage,” Open Biomed Eng J 3, 13–20 (2009).

[70] M. Huber, S. Trattnig, and F. Lintner, “Anatomy, biochemistry, andphysiology of articular cartilage,” Invest Radiol 35, 573–80 (2000).

[71] E. B. Hunziker, T. M. Quinn, and H. J. Hauselmann, “Quantitativestructural organization of normal adult human articular cartilage,”Osteoarthritis Cartilage 10, 564–72 (2002).

[72] M. M. Hyttinen, J. Holopainen, P. R. van Weeren, E. C. Firth, H. J.Helminen, and P. A. Brama, “Changes in collagen fibril network or-ganization and proteoglycan distribution in equine articular cartilageduring maturation and growth,” J Anat 215, 584–91 (2009).

[73] H. Isaksson, J. Töyräs, M. Hakulinen, A. S. Aula, I. Tamminen,P. Julkunen, H. Kröger, and J. S. Jurvelin, “Structural parametersof normal and osteoporotic human trabecular bone are affecteddifferently by microCT image resolution,” Osteoporos Int 22, 167–77(2011).

[74] D. W. Jackson, P. A. Lalor, H. M. Aberman, and T. M. Simon, “Spon-taneous repair of full-thickness defects of articular cartilage in a goatmodel. A preliminary study,” J Bone Joint Surg Am 83-A, 53–64 (2001).

[75] N. S. Joshi, P. N. Bansal, R. C. Stewart, B. D. Snyder, and M. W.Grinstaff, “Effect of contrast agent charge on visualization of artic-ular cartilage using computed tomography: exploiting electrostaticinteractions for improved sensitivity,” J Am Chem Soc 131, 13234–5(2009).

[76] P. Julkunen, W. Wilson, H. Isaksson, J. S. Jurvelin, W. Herzog, andR. K. Korhonen, “A review of the combination of experimentalmeasurements and fibril-reinforced modeling for investigation ofarticular cartilage and chondrocyte response to loading,” ComputMath Methods Med 2013, 326150 (2013).



[77] E. Kaleva, T. Viren, S. Saarakkala, J. Sahlman, J. Sirola, J. Puhakka,T. Paatela, H. Kröger, I. Kiviranta, J. S. Jurvelin, and J. Töyräs,“Arthroscopic Ultrasound Assessment of Articular Cartilage in theHuman Knee Joint: A Potential Diagnostic Method,” Cartilage 2,246–253 (2011).

[78] R. J. Kalke, G. A. Di Primio, and M. E. Schweitzer, “MR and CTarthrography of the knee,” Semin Musculoskelet Radiol 16, 57–68(2012).

[79] A. S. Kallioniemi, J. S. Jurvelin, M. T. Nieminen, M. J. Lammi, andJ. Töyräs, “Contrast agent enhanced pQCT of articular cartilage,”Phys Med Biol 52, 1209–19 (2007).

[80] J. H. Kellgren and J. S. Lawrence, “Radiological assessment of osteo-arthrosis,” Ann Rheum Dis 16, 494–502 (1957).

[81] G. E. Kempson, H. Muir, C. Pollard, and M. Tuke, “The tensileproperties of the cartilage of human femoral condyles related to thecontent of collagen and glycosaminoglycans,” Biochim Biophys Acta297, 456–72 (1973).

[82] K. Kiraly, M. Lammi, J. Arokoski, T. Lapvetelainen, M. Tammi,H. Helminen, and I. Kiviranta, “Safranin O reduces loss of gly-cosaminoglycans from bovine articular cartilage during histologicalspecimen preparation,” Histochem J 28, 99–107 (1996).

[83] I. Kiviranta, J. Jurvelin, M. Tammi, A. M. Saamanen, and H. J. Helmi-nen, “Microspectrophotometric quantitation of glycosaminoglycansin articular cartilage sections stained with Safranin O,” Histochemistry82, 249–55 (1985).

[84] P. Kiviranta, J. Töyräs, M. T. Nieminen, M. S. Laasanen, S. Saarakkala,H. J. Nieminen, M. J. Nissi, and J. S. Jurvelin, “Comparison ofnovel clinically applicable methodology for sensitive diagnostics ofcartilage degeneration,” Eur Cell Mater 13, 46–55 (2007).

[85] H. Kokkonen, J.-S. Suomalainen, A. Joukainen, H. Kröger, J. Sirola,J. Jurvelin, J. Salo, and J. Töyräs, “In vivo diagnostics of humanknee cartilage lesions using delayed CBCT arthrography,” Submitted(2013).

[86] H. T. Kokkonen, A. S. Aula, H. Kröger, J. Suomalainen, E. Lammen-tausta, E. Mervaala, J. Jurvelin, and J. Töyräs, “Delayed computedtomography arthrography of human knee cartilage in vivo,” Cartilage3, 334–41 (2012).


References

[87] H. T. Kokkonen, J. S. Jurvelin, V. Tiitu, and J. Töyräs, “Detectionof mechanical injury of articular cartilage using contrast enhancedcomputed tomography,” Osteoarthritis Cartilage 19, 295–301 (2011).

[88] H. T. Kokkonen, J. Mäkelä, K. A. Kulmala, L. Rieppo, J. S. Jurvelin,V. Tiitu, H. M. Karjalainen, R. K. Korhonen, V. Kovanen, and J. Töyräs,“Computed tomography detects changes in contrast agent diffusionafter collagen cross-linking typical to natural aging of articular carti-lage,” Osteoarthritis Cartilage 19, 1190–8 (2011).

[89] M. Kongsgaard, V. Kovanen, P. Aagaard, S. Doessing, P. Hansen,A. H. Laursen, N. C. Kaldau, M. Kjaer, and S. P. Magnusson, “Corti-costeroid injections, eccentric decline squat training and heavy slowresistance training in patellar tendinopathy,” Scand J Med Sci Sports19, 790–802 (2009).

[90] R. K. Korhonen and J. S. Jurvelin, “Compressive and tensile proper-ties of articular cartilage in axial loading are modulated differentlyby osmotic environment,” Med Eng Phys 32, 155–60 (2010).

[91] R. K. Korhonen, M. S. Laasanen, J. Töyräs, R. Lappalainen, H. J.Helminen, and J. S. Jurvelin, “Fibril reinforced poroelastic modelpredicts specifically mechanical behavior of normal, proteoglycandepleted and collagen degraded articular cartilage,” J Biomech 36,1373–9 (2003).

[92] R. K. Korhonen, M. Wong, J. Arokoski, R. Lindgren, H. J. Helminen,E. B. Hunziker, and J. S. Jurvelin, “Importance of the superficialtissue layer for the indentation stiffness of articular cartilage,” MedEng Phys 24, 99–108 (2002).

[93] J. E. Kurkijarvi, L. Mattila, R. O. Ojala, A. I. Vasara, J. S. Jurvelin,I. Kiviranta, and M. T. Nieminen, “Evaluation of cartilage repair inthe distal femur after autologous chondrocyte transplantation usingT2 relaxation time and dGEMRIC,” Osteoarthritis Cartilage 15, 372–8(2007).

[94] J. E. Kurkijarvi, M. J. Nissi, I. Kiviranta, J. S. Jurvelin, and M. T. Niem-inen, “Delayed gadolinium-enhanced MRI of cartilage (dGEMRIC)and T2 characteristics of human knee articular cartilage: topograph-ical variation and relationships to mechanical properties,” MagnReson Med 52, 41–6 (2004).



[95] M. S. Laasanen, J. Töyräs, R. K. Korhonen, J. Rieppo, S. Saarakkala,M. T. Nieminen, J. Hirvonen, and J. S. Jurvelin, “Biomechanicalproperties of knee articular cartilage,” Biorheology 40, 133–140 (2003).

[96] W. M. Lai, J. S. Hou, and V. C. Mow, “A triphasic theory for theswelling and deformation behaviors of articular cartilage,” J BiomechEng 113, 245–58 (1991).

[97] B. A. Lakin, D. J. Grasso, S. S. Shah, R. C. Stewart, P. N. Bansal,J. D. Freedman, M. W. Grinstaff, and B. D. Snyder, “Cationic agentcontrast-enhanced computed tomography imaging of cartilage cor-relates with the compressive modulus and coefficient of friction,”Osteoarthritis Cartilage 21, 60–8 (2013).

[98] E. Lammentausta, P. Kiviranta, M. J. Nissi, M. S. Laasanen, I. Kivi-ranta, M. T. Nieminen, and J. S. Jurvelin, “T2 relaxation time anddelayed gadolinium-enhanced MRI of cartilage (dGEMRIC) of hu-man patellar cartilage at 1.5 T and 9.4 T: Relationships with tissuemechanical properties,” J Orthop Res 24, 366–74 (2006).

[99] R. C. Lawrence, D. T. Felson, C. G. Helmick, L. M. Arnold, H. Choi,R. A. Deyo, S. Gabriel, R. Hirsch, M. C. Hochberg, G. G. Hunder,J. M. Jordan, J. N. Katz, H. M. Kremers, and F. Wolfe, “Estimatesof the prevalence of arthritis and other rheumatic conditions in theUnited States. Part II,” Arthritis Rheum 58, 26–35 (2008).

[100] H. A. Leddy and F. Guilak, “Site-specific molecular diffusion inarticular cartilage measured using fluorescence recovery after photo-bleaching,” Ann Biomed Eng 31, 753–60 (2003).

[101] H. A. Leddy, M. A. Haider, and F. Guilak, “Diffusional anisotropyin collagenous tissues: fluorescence imaging of continuous pointphotobleaching,” Biophys J 91, 311–6 (2006).

[102] L. M. Lesperance, M. L. Gray, and D. Burstein, “Determination offixed charge density in cartilage using nuclear magnetic resonance,”J Orthop Res 10, 1–13 (1992).

[103] L. P. Li, J. Soulhat, M. D. Buschmann, and A. Shirazi-Adl, “Nonlinearanalysis of cartilage in unconfined ramp compression using a fibrilreinforced poroelastic model,” Clin Biomech (Bristol, Avon) 14, 673–82(1999).

[104] W. Li, R. Scheidegger, Y. Wu, R. R. Edelman, M. Farley, N. Krishnan,D. Burstein, and P. V. Prasad, “Delayed contrast-enhanced MRI of


References

cartilage: Comparison of nonionic and ionic contrast agents,” MagnReson Med 64, 1267–73 (2010).

[105] K. Lindblom, “The Arthrographic Appearance of the Ligaments ofthe Knee Joint,” Acta Radiol 19, 582–600 (1938).

[106] A. Maroudas, “Physicochemical properties of cartilage in the lightof ion exchange theory,” Biophys J 8, 575–95 (1968).

[107] A. Maroudas, “Distribution and diffusion of solutes in articularcartilage,” Biophys J 10, 365–79 (1970).

[108] A. Maroudas, “Biophysical chemistry of cartilaginous tissues withspecial reference to solute and fluid transport,” Biorheology 12, 233–48.(1975).

[109] A. Maroudas, H. Muir, and J. Wingham, “The correlation of fixednegative charge with glycosaminoglycan content of human articularcartilage,” Biochim Biophys Acta 177, 492–500 (1969).

[110] A. I. Maroudas, “Balance between swelling pressure and colla-gen tension in normal and degenerate cartilage,” Nature 260, 808–9.(1976).

[111] J. A. Martin and J. A. Buckwalter, “Roles of articular cartilage agingand chondrocyte senescence in the pathogenesis of osteoarthritis,”Iowa Orthop J 21, 1–7 (2001).

[112] S. A. Mazzuca, K. D. Brandt, K. A. Lane, and B. P. Katz, “Kneepain reduces joint space width in conventional standing anteroposte-rior radiographs of osteoarthritic knees,” Arthritis Rheum 46, 1223–7(2002).

[113] C. W. McCutchen, “Cartilage is poroelastic, not viscoelastic (includ-ing an exact theorem about strain energy and viscous loss, and anorder of magnitude relation for equilibration time),” J Biomech 15,325–7 (1982).

[114] C. A. McKenzie, A. Williams, P. V. Prasad, and D. Burstein, “Three-dimensional delayed gadolinium-enhanced MRI of cartilage (dGEM-RIC) at 1.5T and 3.0T,” J Magn Reson Imaging 24, 928–33 (2006).

[115] J. Mizrahi, A. Maroudas, Y. Lanir, I. Ziv, and T. J. Webber, “The"instantaneous" deformation of cartilage: effects of collagen fiberorientation and osmotic stress,” Biorheology 23, 311–30 (1986).



[116] I. Moller, D. Bong, E. Naredo, E. Filippucci, I. Carrasco, C. Moragues,and A. Iagnocco, “Ultrasound in the study and monitoring ofosteoarthritis,” Osteoarthritis Cartilage 16 Suppl 3, S4–7 (2008).

[117] V. C. Mow, D. C. Fithian, and M. A. Kelly, “Fundamentals of articularcartilage and meniscus biomechanics,” in Articular cartilage and kneejoint function: basic science and arthroscopy (Raven Press Ltd., NewYork, 1990), pp. 1-18.

[118] V. C. Mow and X. Guo, “Mechano-electrochemical properties ofarticular cartilage: their inhomogeneities and anisotropies,” AnnuRev Biomed Eng 4, 175–209 (2002).

[119] V. C. Mow, M. H. Holmes, and W. M. Lai, “Fluid transport andmechanical properties of articular cartilage: a review,” J Biomech 17,377–94 (1984).

[120] V. C. Mow, S. C. Kuei, W. M. Lai, and C. G. Armstrong, “Biphasiccreep and stress relaxation of articular cartilage in compression:Theory and experiments,” J Biomech Eng 102, 73–84 (1980).

[121] V. C. Mow, A. Ratcliffe, and A. R. Poole, “Cartilage and diarthro-dial joints as paradigms for hierarchical materials and structures,”Biomaterials 13, 67–97 (1992).

[122] V. C. Mow, W. Zhu, and A. Ratcliffe, “Structure and function ofarticular cartilage and meniscus,” in Basic orthopaedic biomechanics,V. C. Mow and W. C. Hayes, eds. (Raven Press, Ltd, New York, 1991),pp. 143-198.

[123] H. Muir, “The chondrocyte, architect of cartilage. Biomechanics,structure, function and molecular biology of cartilage matrix macro-molecules,” Bioessays 17, 1039–48 (1995).

[124] H. Muir, P. Bullough, and A. Maroudas, “The distribution of col-lagen in human articular cartilage with some of its physiologicalimplications,” J Bone Joint Surg Br 52, 554–63. (1970).

[125] J. Multanen, E. Rauvala, E. Lammentausta, R. Ojala, I. Kiviranta,A. Hakkinen, M. T. Nieminen, and A. Heinonen, “Reproducibilityof imaging human knee cartilage by delayed gadolinium-enhancedMRI of cartilage (dGEMRIC) at 1.5 Tesla,” Osteoarthritis Cartilage 17,559–64 (2009).


References

[126] National Institute of Arthritis and Musculoskeletal and Skin Dis-eases, “Handout on Health: Osteoarthritis, NIH Publication No.10-4617,” http://www.niams.nih.gov/health_info/Osteoarthritis/default.asp (2010).

[127] National Institutes of Health, “NIH Consensus Statement on totalknee replacement,” NIH Consens State Sci Statements 20, 1–34 (2003).

[128] P. Neuman, J. Tjornstrand, J. Svensson, C. Ragnarsson, H. Roos,M. Englund, C. J. Tiderius, and L. E. Dahlberg, “Longitudinalassessment of femoral knee cartilage quality using contrast enhancedMRI (dGEMRIC) in patients with anterior cruciate ligament injury–comparison with asymptomatic volunteers,” Osteoarthritis Cartilage19, 977–83 (2011).

[129] M. Nieminen, J. Rieppo, J. Silvennoinen, J. Töyräs, J. Hakumäki,M. Hyttinen, H. Helminen, and J. Jurvelin, “Spatial assessmentof articular cartilage proteoglycans with Gd-DTPA -enhanced T(1)imaging,” Magn Reson Med 48, 640–648 (2002).

[130] M. T. Nieminen, J. Töyräs, M. S. Laasanen, J. Silvennoinen, H. J.Helminen, and J. S. Jurvelin, “Prediction of biomechanical propertiesof articular cartilage with quantitative magnetic resonance imaging,”J Biomech 37, 321–8 (2004).

[131] M. J. Nissi, J. Töyräs, M. S. Laasanen, J. Rieppo, S. Saarakkala,R. Lappalainen, J. S. Jurvelin, and M. T. Nieminen, “Proteoglycanand collagen sensitive MRI evaluation of normal and degeneratedarticular cartilage,” J Orthop Res 22, 557–64 (2004).

[132] T. Nojiri, N. Watanabe, T. Namura, W. Narita, K. Ikoma, T. Sug-inoshita, H. Takamiya, H. Komiyama, H. Ito, T. Nishimura, andT. Kubo, “Utility of delayed gadolinium-enhanced MRI (dGEM-RIC) for qualitative evaluation of articular cartilage of patellofemoraljoint,” Knee Surg Sports Traumatol Arthrosc 14, 718–23 (2006).

[133] A. Oloyede and N. D. Broom, “Is classical consolidation theoryapplicable to articular cartilage deformation?,” Clin Biomech 6, 206–212 (1991).

[134] H. Owman, C. J. Tiderius, P. Neuman, F. Nyquist, and L. E. Dahlberg,“Association between findings on delayed gadolinium-enhanced mag-netic resonance imaging of cartilage and future knee osteoarthritis,”Arthritis Rheum 58, 1727–30 (2008).



[135] A. W. Palmer, R. E. Guldberg, and M. E. Levenston, “Analysis ofcartilage matrix fixed charge density and three-dimensional mor-phology via contrast-enhanced microcomputed tomography,” ProcNatl Acad Sci USA 103, 19255–60 (2006).

[136] L. O. Parker and A. A. Delorimier, “Double contrast visualization ofjoints,” Calif Med 78, 424–7 (1953).

[137] T. J. Perlewitz, V. M. Haughton, r. Riley, L. H., C. Nguyen-Minh, andV. George, “Effect of molecular weight on the diffusion of contrastmedia into cartilage,” Spine (Phila Pa 1976) 22, 2707–10 (1997).

[138] T. M. Piscaer, G. J. van Osch, J. A. Verhaar, and H. Weinans, “Imagingof experimental osteoarthritis in small animal models,” Biorheology45, 355–64 (2008).

[139] T. M. Piscaer, J. H. Waarsing, N. Kops, P. Pavljasevic, J. A. Verhaar,G. J. van Osch, and H. Weinans, “In vivo imaging of cartilagedegeneration using microCT-arthrography,” Osteoarthritis Cartilage16, 1011–7 (2008).

[140] K. Pridie, “A method of resurfacing osteoarthritic knee joints.,”Journal of Bone and Joint Surgery Br 41, 618–619 (1959).

[141] P. Qvist, A. C. Bay-Jensen, C. Christiansen, E. B. Dam, P. Pastoureau,and M. A. Karsdal, “The disease modifying osteoarthritis drug(DMOAD): Is it in the horizon?,” Pharmacol Res 58, 1–7 (2008).

[142] E. Radin and R. Rose, “Role of subchondral bone in the initiationand progression of cartilage damage,” Clin Orthop Relat Res. 213,34–40 (1986).

[143] M. Reiser, P. M. Karpf, and P. Bernett, “Diagnosis of chondromalaciapatellae using CT arthrography,” Eur J Radiol 2, 181–6 (1982).

[144] J. Rieppo, J. Hallikainen, J. S. Jurvelin, I. Kiviranta, H. J. Helminen,and M. M. Hyttinen, “Practical considerations in the use of polarizedlight microscopy in the analysis of the collagen network in articularcartilage,” Microsc Res Tech 71, 279–87 (2008).

[145] L. Rieppo, S. Saarakkala, T. Närhi, J. Holopainen, M. Lammi, H. J.Helminen, J. S. Jurvelin, and J. Rieppo, “Quantitative analysis of spa-tial proteoglycan content in articular cartilage with Fourier transforminfrared imaging spectroscopy: Critical evaluation of analysis meth-ods and specificity of the parameters,” Microsc Res Tech 73, 503–12(2010).


References

[146] J. J. Rodrigo, J. R. Steadman, and J. F. Silliman, “Improvementin full thickness chondral defect healing in the human knee afterdebridement and microfracture using continuous passive motion.,”Am J Knee Surg. 7, 109–116 (1994).

[147] E. J. Roebuck, “Double contrast knee arthrography. Some new pointsof technique including the use of Dimer X,” Clin Radiol 28, 247–57(1977).

[148] L. E. Romans, “Contrast agents,” in Computed tomography for technol-ogists: exam review (Lippincott Williams & Wilkins, Philadelphia, Pa.;London, 2010), pp. 52-64.

[149] V. Roth and V. C. Mow, “The intrinsic tensile behavior of the matrixof bovine articular cartilage and its variation with age,” J Bone JointSurg Am 62, 1102–17 (1980).

[150] P. Ruegsegger, B. Koller, and R. Muller, “A microtomographic systemfor the nondestructive evaluation of bone architecture,” Calcif TissueInt 58, 24–9 (1996).

[151] S. Saarakkala, P. Waris, V. Waris, I. Tarkiainen, E. Karvanen, J. Aarnio,and J. M. Koski, “Diagnostic performance of knee ultrasonographyfor detecting degenerative changes of articular cartilage,” Osteoarthri-tis Cartilage 20, 376–81 (2012).

[152] E. N. Salo, M. J. Nissi, K. A. Kulmala, V. Tiitu, J. Töyräs, and M. T.Nieminen, “Diffusion of Gd-DTPA(2) into articular cartilage,” Os-teoarthritis Cartilage 20, 117–26 (2012).

[153] J. T. Samosky, D. Burstein, W. Eric Grimson, R. Howe, S. Martin, andM. L. Gray, “Spatially-localized correlation of dGEMRIC-measuredGAG distribution and mechanical stiffness in the human tibialplateau,” J Orthop Res 23, 93–101 (2005).

[154] SCANCO Medical, “XtremeCT : Clinical microCT : Systems &Solutions : SCANCO Medical AG,” http://www.scanco.ch/index.php?id=xtremect (2013).

[155] D. R. Sell and V. M. Monnier, “Isolation, purification and partialcharacterization of novel fluorophores from aging human insolublecollagen-rich tissue,” Connect Tissue Res 19, 77–92 (1989).

[156] L. A. Setton, W. Zhu, and V. C. Mow, “The biphasic poroviscoelasticbehavior of articular cartilage: role of the surface zone in governingthe compressive behavior,” J Biomech 26, 581–92 (1993).



[157] E. M. Shapiro, A. Borthakur, J. H. Kaufman, J. S. Leigh, and R. Reddy,“Water distribution patterns inside bovine articular cartilage as visu-alized by 1H magnetic resonance imaging,” Osteoarthritis Cartilage 9,533–8 (2001).

[158] L. Sharma, D. Kapoor, and S. Issa, “Epidemiology of osteoarthritis:an update,” Curr Opin Rheumatol 18, 147–56 (2006).

[159] T. Shirai, M. Kobayashi, S. Nakamura, R. Arai, K. Nishitani, T. Satake,L. E. Dahlberg, H. Kuroki, Y. Nakagawa, T. Okada, K. Togashi, andT. Nakamura, “Longitudinal evaluation of cartilage after osteochon-dral autogenous transfer with delayed gadolinium-enhanced MRI ofthe cartilage (dGEMRIC),” J Orthop Res 30, 221–5 (2012).

[160] M. Siebelt, J. van Tiel, J. H. Waarsing, T. M. Piscaer, M. van Straten,R. Booij, M. L. Dijkshoorn, G. J. Kleinrensink, J. A. Verhaar, G. P.Krestin, H. Weinans, and E. H. Oei, “Clinically applied CT arthrog-raphy to measure the sulphated glycosaminoglycan content of carti-lage,” Osteoarthritis Cartilage 19, 1183–9 (2011).

[161] M. Siebelt, J. H. Waarsing, N. Kops, T. M. Piscaer, J. A. Verhaar, E. H.Oei, and H. Weinans, “Quantifying osteoarthritic cartilage changesaccurately using in vivo microCT arthrography in three etiologicallydistinct rat models,” J Orthop Res 29, 1788–94 (2011).

[162] T. S. Silvast, J. Jurvelin, V. Tiitu, T. M. Quinn, and J. Töyräs, “BathConcentration of Anionic Contrast Agents Does Not Affect TheirDiffusion and Distribution in Articular Cartilage In Vitro,” Cartilage4, 42–51 (2013).

[163] T. S. Silvast, J. S. Jurvelin, A. S. Aula, M. J. Lammi, and J. Töyräs,“Contrast agent-enhanced computed tomography of articular carti-lage: association with tissue composition and properties,” Acta Radiol50, 78–85 (2009).

[164] T. S. Silvast, J. S. Jurvelin, M. J. Lammi, and J. Töyräs, “pQCTstudy on diffusion and equilibrium distribution of iodinated anioniccontrast agent in human articular cartilage — associations to matrixcomposition and integrity,” Osteoarthritis Cartilage 17, 26–32 (2009).

[165] T. S. Silvast, H. T. Kokkonen, J. S. Jurvelin, T. M. Quinn, M. T. Niemi-nen, and J. Töyräs, “Diffusion and near-equilibrium distribution ofMRI and CT contrast agents in articular cartilage,” Phys Med Biol 54,6823–36 (2009).


References

[166] M. A. Soltz and G. A. Ateshian, “A Conewise Linear Elasticitymixture model for the analysis of tension- compression nonlinearityin articular cartilage,” J Biomech Eng 122, 576–86. (2000).

[167] G. Spahn, H. M. Klinger, M. Baums, M. Hoffmann, H. Plettenberg,A. Kroker, and G. O. Hofmann, “Near-infrared spectroscopy forarthroscopic evaluation of cartilage lesions: results of a blinded,prospective, interobserver study,” Am J Sports Med 38, 2516–21 (2010).

[168] G. Spahn, H. M. Klinger, and G. O. Hofmann, “How valid is thearthroscopic diagnosis of cartilage lesions? Results of an opinion sur-vey among highly experienced arthroscopic surgeons,” Arch OrthopTrauma Surg 129, 1117–21 (2009).

[169] D. G. Steele and C. A. Bramblett, The anatomy and biology of the humanskeleton, 1st ed. ed. (Texas A&M University Press, College Station,1988).

[170] R. C. Stewart, P. N. Bansal, V. Entezari, H. Lusic, R. M. Nazarian,B. D. Snyder, and M. W. Grinstaff, “Contrast-enhanced CT with aHigh-Affinity Cationic Contrast Agent for Imaging ex Vivo Bovine,Intact ex Vivo Rabbit, and in Vivo Rabbit Cartilage,” Radiology 266,141–50 (2013).

[171] J. J. Stubendorff, E. Lammentausta, A. Struglics, L. Lindberg,D. Heinegard, and L. E. Dahlberg, “Is cartilage sGAG content relatedto early changes in cartilage disease? Implications for interpretationof dGEMRIC,” Osteoarthritis Cartilage 20, 396–404 (2012).

[172] J. K. Suh and S. Bai, “Finite element formulation of biphasic porovis-coelastic model for articular cartilage,” J Biomech Eng 120, 195–201(1998).

[173] C. Tiderius, M. Hori, A. Williams, L. Sharma, P. V. Prasad, M. Finnell,C. McKenzie, and D. Burstein, “dGEMRIC as a function of BMI,”Osteoarthritis Cartilage 14, 1091–7 (2006).

[174] C. J. Tiderius, R. Jessel, Y. J. Kim, and D. Burstein, “Hip dGEMRICin asymptomatic volunteers and patients with early osteoarthritis:the influence of timing after contrast injection,” Magn Reson Med 57,803–5 (2007).

[175] C. J. Tiderius, L. E. Olsson, H. de Verdier, P. Leander, O. Ekberg, andL. Dahlberg, “Gd-DTPA2)-enhanced MRI of femoral knee cartilage:a dose-response study in healthy volunteers,” Magn Reson Med 46,1067–71 (2001).



[176] C. J. Tiderius, L. E. Olsson, P. Leander, O. Ekberg, and L. Dahlberg,“Delayed gadolinium-enhanced MRI of cartilage (dGEMRIC) in earlyknee osteoarthritis,” Magn Reson Med 49, 488–92 (2003).

[177] C. J. Tiderius, J. Svensson, P. Leander, T. Ola, and L. Dahlberg,“dGEMRIC (delayed gadolinium-enhanced MRI of cartilage) indi-cates adaptive capacity of human knee cartilage,” Magn Reson Med51, 286–90 (2004).

[178] C. J. Tiderius, J. Tjornstrand, P. Akeson, K. Sodersten, L. Dahlberg,and P. Leander, “Delayed gadolinium-enhanced MRI of cartilage(dGEMRIC): intra- and interobserver variability in standardizeddrawing of regions of interest,” Acta Radiol 45, 628–34 (2004).

[179] P. A. Torzilli, “Effects of temperature, concentration and articularsurface removal on transient solute diffusion in articular cartilage,”Med Biol Eng Comput 31 Suppl, S93–8 (1993).

[180] P. A. Torzilli, T. C. Adams, and R. J. Mis, “Transient solute diffusionin articular cartilage,” J Biomech 20, 203–14 (1987).

[181] P. A. Torzilli, J. M. Arduino, J. D. Gregory, and M. Bansal, “Effectof proteoglycan removal on solute mobility in articular cartilage,” JBiomech 30, 895–902 (1997).

[182] P. A. Torzilli, D. A. Grande, and J. M. Arduino, “Diffusive propertiesof immature articular cartilage,” J Biomed Mater Res 40, 132–8 (1998).

[183] S. Trattnig, T. C. Mamisch, K. Pinker, S. Domayer, P. Szomolanyi,S. Marlovits, F. Kutscha-Lissberg, and G. H. Welsch, “Differentiat-ing normal hyaline cartilage from post-surgical repair tissue usingfast gradient echo imaging in delayed gadolinium-enhanced MRI(dGEMRIC) at 3 Tesla,” Eur Radiol 18, 1251–9 (2008).

[184] S. Trattnig, S. Marlovits, S. Gebetsroither, P. Szomolanyi, G. H.Welsch, E. Salomonowitz, A. Watanabe, M. Deimling, and T. C.Mamisch, “Three-dimensional delayed gadolinium-enhanced MRIof cartilage (dGEMRIC) for in vivo evaluation of reparative cartilageafter matrix-associated autologous chondrocyte transplantation at3.0T: Preliminary results,” J Magn Reson Imaging 26, 974–82 (2007).

[185] S. Trattnig, V. Mlynarik, M. Breitenseher, M. Huber, A. Zembsch,T. Rand, and H. Imhof, “MRI visualization of proteoglycan depletionin articular cartilage via intravenous administration of Gd-DTPA,”Magn Reson Imaging 17, 577–83 (1999).


References

[186] M. J. Turunen, J. Töyräs, M. J. Lammi, J. S. Jurvelin, and R. K.Korhonen, “Hyperosmolaric contrast agents in cartilage tomographymay expose cartilage to overload-induced cell death,” J Biomech 45,497–503 (2012).

[187] J. Töyräs, J. Rieppo, M. T. Nieminen, H. J. Helminen, and J. S. Jurvelin,“Characterization of enzymatically induced degradation of articularcartilage using high frequency ultrasound,” Phys Med Biol 44, 2723–33 (1999).

[188] U.S. Food and Drug Administration, “FDA Drug Safety Newsletter:Questions and Answers on Gadolinium-Based Contrast Agents,”http://www.fda.gov/Drugs/DrugSafety/DrugSafetyNewsletter/ucm142889.htm (2009).

[189] J. van Tiel, M. Siebelt, J. H. Waarsing, T. M. Piscaer, M. van Straten,R. Booij, M. L. Dijkshoorn, G. J. Kleinrensink, J. A. Verhaar, G. P.Krestin, H. Weinans, and E. H. Oei, “CT arthrography of the hu-man knee to measure cartilage quality with low radiation dose,”Osteoarthritis Cartilage 20, 678–85 (2012).

[190] B. C. Vande Berg, F. E. Lecouvet, P. Poilvache, B. Maldague, andJ. Malghem, “Spiral CT arthrography of the knee: technique andvalue in the assessment of internal derangement of the knee,” EurRadiol 12, 1800–10 (2002).

[191] A. I. Vasara, M. M. Hyttinen, M. J. Lammi, P. E. Lammi, T. K. Langsjo,A. Lindahl, L. Peterson, M. Kellomaki, Y. T. Konttinen, H. J. Helmi-nen, and I. Kiviranta, “Subchondral bone reaction associated withchondral defect and attempted cartilage repair in goats,” Calcif TissueInt 74, 107–14 (2004).

[192] A. I. Vasara, M. M. Hyttinen, O. Pulliainen, M. J. Lammi, J. S. Jurvelin,L. Peterson, A. Lindahl, H. J. Helminen, and I. Kiviranta, “Immatureporcine knee cartilage lesions show good healing with or withoutautologous chondrocyte transplantation,” Osteoarthritis Cartilage 14,1066–74 (2006).

[193] M. Venn and A. Maroudas, “Chemical composition and swellingof normal and osteoarthrotic femoral head cartilage. I. Chemicalcomposition,” Ann Rheum Dis 36, 121–9 (1977).

[194] N. Verzijl, J. DeGroot, Z. C. Ben, O. Brau-Benjamin, A. Maroudas,R. A. Bank, J. Mizrahi, C. G. Schalkwijk, S. R. Thorpe, J. W. Baynes,



J. W. Bijlsma, F. P. Lafeber, and J. M. TeKoppele, “Crosslinkingby advanced glycation end products increases the stiffness of thecollagen network in human articular cartilage: a possible mechanismthrough which age is a risk factor for osteoarthritis,” Arthritis Rheum46, 114–23 (2002).

[195] N. Verzijl, J. DeGroot, E. Oldehinkel, R. A. Bank, S. R. Thorpe, J. W.Baynes, M. T. Bayliss, J. W. Bijlsma, F. P. Lafeber, and J. M. Tekoppele,“Age-related accumulation of Maillard reaction products in humanarticular cartilage collagen,” Biochem J 350 Pt 2, 381–7 (2000).

[196] T. Viren, S. Saarakkala, E. Kaleva, H. J. Nieminen, J. S. Jurvelin,and J. Töyräs, “Minimally invasive ultrasound method for intra-articular diagnostics of cartilage degeneration,” Ultrasound Med Biol35, 1546–54 (2009).

[197] A. Watanabe, C. Boesch, S. E. Anderson, W. Brehm, and P. Mainil Var-let, “Ability of dGEMRIC and T2 mapping to evaluate cartilage repairafter microfracture: a goat study,” Osteoarthritis Cartilage 17, 1341–9(2009).

[198] A. Watanabe, Y. Wada, T. Obata, T. Ueda, M. Tamura, H. Ikehira,and H. Moriya, “Delayed gadolinium-enhanced MR to determineglycosaminoglycan concentration in reparative cartilage after autolo-gous chondrocyte implantation: preliminary results,” Radiology 239,201–8 (2006).

[199] E. Wiener, J. Hodler, and C. W. Pfirrmann, “Delayed gadolinium-enhanced MRI of cartilage (dGEMRIC) of cadaveric shoulders: com-parison of contrast dynamics in hyaline and fibrous cartilage afterintraarticular gadolinium injection,” Acta Radiol 50, 86–92 (2009).

[200] E. Wiener, M. Settles, G. Weirich, C. Schmidt, and G. Diederichs,“The influence of collagen network integrity on the accumulation ofgadolinium-based MR contrast agents in articular cartilage,” FortschrRöntgenstr 183, 226–232 (2011).

[201] E. Wiener, K. Woertler, G. Weirich, E. J. Rummeny, and M. Settles,“Contrast enhanced cartilage imaging: Comparison of ionic andnon-ionic contrast agents,” Eur J Radiol 63, 110–9 (2007).

[202] A. Williams, A. Gillis, C. McKenzie, B. Po, L. Sharma, L. Micheli,B. McKeon, and D. Burstein, “Glycosaminoglycan distribution in car-tilage as determined by delayed gadolinium-enhanced MRI of carti-


References

lage (dGEMRIC): potential clinical applications,” AJR Am J Roentgenol182, 167–72 (2004).

[203] A. Williams, B. Mikulis, N. Krishnan, M. Gray, C. McKenzie, andD. Burstein, “Suitability of T(1Gd) as the dGEMRIC index at 1.5Tand 3.0T,” Magn Reson Med 58, 830–4 (2007).

[204] A. Williams, L. Sharma, C. A. McKenzie, P. V. Prasad, and D. Burstein,“Delayed gadolinium-enhanced magnetic resonance imaging of carti-lage in knee osteoarthritis: findings at different radiographic stagesof disease and relationship to malalignment,” Arthritis Rheum 52,3528–35 (2005).

[205] W. Wilson, C. C. van Donkelaar, B. van Rietbergen, and R. Huiskes,“A fibril-reinforced poroviscoelastic swelling model for articular car-tilage,” J Biomech 38, 1195–204 (2005).

[206] W. Wilson, C. C. van Donkelaar, B. van Rietbergen, K. Ito, andR. Huiskes, “Stresses in the local collagen network of articularcartilage: a poroviscoelastic fibril-reinforced finite element study,” JBiomech 37, 357–66 (2004).

[207] W. Wilson, C. C. van Donkelaar, R. van Rietbergen, and R. Huiskes,“The role of computational models in the search for the mechanicalbehavior and damage mechanisms of articular cartilage,” Med EngPhys 27, 810–26 (2005).

[208] Y. Xia, S. Zheng, and A. Bidthanapally, “Depth-dependent pro-files of glycosaminoglycans in articular cartilage by microMRI andhistochemistry,” J Magn Reson Imaging 28, 151–7 (2008).

[209] L. Xie, A. S. Lin, R. E. Guldberg, and M. E. Levenston, “Nondestruc-tive assessment of sGAG content and distribution in normal anddegraded rat articular cartilage via EPIC-microCT,” OsteoarthritisCartilage 18, 65–72 (2010).

[210] K. Zheng, S. D. Martin, C. H. Rashidifard, B. Liu, and M. E. Brezin-ski, “In vivo micron-scale arthroscopic imaging of human kneeosteoarthritis with optical coherence tomography: comparison withmagnetic resonance imaging and arthroscopy,” Am J Orthop (BelleMead NJ) 39, 122–5 (2010).

[211] W. Zhu, V. C. Mow, T. J. Koob, and D. R. Eyre, “Viscoelastic shearproperties of articular cartilage and the effects of glycosidase treat-ments,” J Orthop Res 11, 771–81 (1993).


Publications of the University of Eastern Finland Dissertations in Forestry and Natural Sciences

Publications of the University of Eastern Finland

Dissertations in Forestry and Natural Sciences

isbn 978-952-61-1158-2

Katariina Nissinen

Diffusion of X-ray Contrast Agents in Intact and Repaired Articular Cartilage

Early degenerative changes

of articular cartilage are not

diagnosed sensitively enough with

conventional imaging techniques.

The diagnostic potential of contrast

enhanced computed tomography

(CECT) was investigated in this

thesis in vitro and in vivo. Imaging

the diffusion dynamics of the

contrast agent in articular cartilage

was found to provide valuable

information about the integrity and

composition of cartilage matrix.

Based on the present findings, it

seems that delayed quantitative CT

arthrography (dQCTA i.e. CECT in

vivo) is a potentially useful method

in the quantitative diagnostic

imaging of articular cartilage.

dissertatio

ns | 110 | K

atar

iina N

issinen

| Diffu

sion

of X-ra

y Co

ntra

st Agen

ts in In

tact a

nd

Rep

aired

Articu

lar C

artila

ge

Katariina NissinenDiffusion of X-ray Contrast

Agents in Intact and Repaired Articular Cartilage

| 110 | Katariina Nissinen | Katariina Nissinen Diffusion...

Documents

Transcript of | 110 | Katariina Nissinen | Katariina Nissinen Diffusion...