UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Effects of alendronate-immobilized calcium phosphate coating on bone growth into porous tantalum : a… Hu, Youxin 2007

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_2007-0119a.pdf [ 12.42MB ]
Metadata
JSON: 831-1.0078462.json
JSON-LD: 831-1.0078462-ld.json
RDF/XML (Pretty): 831-1.0078462-rdf.xml
RDF/JSON: 831-1.0078462-rdf.json
Turtle: 831-1.0078462-turtle.txt
N-Triples: 831-1.0078462-rdf-ntriples.txt
Original Record: 831-1.0078462-source.json
Full Text
831-1.0078462-fulltext.txt
Citation
831-1.0078462.ris

Full Text

EFFECTS OF ALENDRONATE-IMMOBILIZED CALCIUM PHOSPHATE COATING ON BONE GROWTH INTO POROUS TANTALUM —A GAP MODEL ANIMAL STUDY by  YOUXIN HU M.A.Sc. in Materials Science & Engineering, Northeast Univ. of Tech. China, 1990 B.A.Sc. in Materials Science & Engineering, Northeast Univ. of Tech. China, 1985  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  MASTER O F APPLIED SCIENCE in THE FACULTY OF GRADUATE STUDIES (Materials Engineering)  THE UNIVERSITY OF BRITISH COLUMBIA April 2007  © Youxin Hu, 2007  ABSTRACT Porous tantalum has been shown to be a promising orthopaedic implant material because o f its similarity to bone in both mechanical properties and its three-dimensional porous structure. However, in some circumstances, bone quality or quantity is insufficient to allow adequate bone ingrowth. Alendronate, one o f the bisphosphonate families, affects the activities o f bone cells and enhances bone formation. In this thesis, we hypothesized that the addition o f alendronate could increase the osteoconductivity and bone-ingrowth o f porous tantalum and overcome the challenges o f bone-implant gaps.  To facilitate local delivery o f alendronate, a micro-porous calcium phosphate coating was deposited onto the tantalum surface by an electrolytic deposition technique, which was followed by alendronate adsorption. Coating structures and morphologies were confirmed by scanning electron microscopy, X-ray diffraction, and Fourier transform infrared spectroscopy. The presence o f alendronate was confirmed by high performance liquid chromatography.  To study the effects o f alendronate-immobilized calcium phosphate coating on bone reaction to porous implants, an animal gap model, with a fixed gap o f 0.6 m m between implants and bone, was developed. Three types o f surfaces, which were non-coating (Ta), calcium phosphate coating (Ta-CaP), and alendronate-immobilized calcium phosphate coating (Ta-CaP-AIN), were compared. T w o fluorochromes were adopted to track the front o f bone formation. After four weeks o f healing and following standard histology techniques, the implants were analyzed with backscattered electron microscopy and fluorescent optical microscopy for bone-implant interactions.  ii  The relative volume increase o f gap filling, bone ingrowth and total bone formation were 124 % (2.24-fold), 232% (3.32-fold) and 170% (2.7-fold) respectively in T a - C a P - A L N compared with T a controls. The contact length o f newly formed bone on porous tantalum was increased by 700% (8-fold) i n T a - C a P - A L N compared with T a plugs, suggesting enhanced osteoconductivity o f T a - C a P - A L N  implants. The bone formation mechanism  analysis found that bone growth initiated on both surfaces o f the T a - C a P - A L N implants and host bone, while little bone initiation on the Ta implant surface was detected.  These  significant enhancements o f T a - C a P ^ A L N may have direct applications in orthopaedics. For revision arthroplasty with insufficient bone stock, the local delivery o f alendronate would enhance biological fixation o f the implant and promote the healing o f bone defects.  iii  TABLE OF CONTENTS ABSTRACT  ii  T A B L E OF CONTENTS  iv  LIST OF T A B L E S  vii  LIST OF FIGURES  viii  LIST OF A B B R E V I A T I O N  xiii  ACKNOWLEDGMENTS  xiv  C H A P T E R 1: I N T R O D U C T I O N  1  C H A P T E R 2: L I T E R A T U R E R E V I E W  4  2.1 Introduction to artificial hip joints  4  2.2.1 Structure o f bone  6  2.2.3 The mechanical properties o f bone  9  2.2.4 Bone formation and bone healing process  10  2.3 Implant materials  11  2.4 The challenges in hip replacement  14  2.5 Current material approaches to improve implant-bone fixation  16  2.6 Calcium phosphate  17  2.7 Deposition o f hydroxyapatite  19  2.7 Effect o f bisphosphonates on bone  22  2.8 Current studies on bisphosphonate-delivery implants  23  2.9 Gap model animal tests  24  2.10 Mechanism o f labeling  25  2.11. t-tests  26  C H A P T E R 3: S C O P E A N D O B J E C T I V E S  29 iv  C H A P T E R 4: M A T E R I A L S A N D M E T H O D S  31  4.1 Implant preparation  31  4.1.1 Porous tantalum  31  4.1.2 Electrolytic deposition ( E L D ) o f calcium phosphate coatings  33  4.1.3 Alendronate immobilization  36  4.1.4 Implant assemblies  37  4.1.5 Sterilization o f implants and surgical tools  39  4.2 H i g h performance liquid chromatography ( H P L C ) tests for drug loading  40  4.3 A n i m a l tests  40  4.3.1 Implantation procedures  42  4.3.2 Bone labeling  43  4.3.3 A n i m a l care after surgery  44  4.3.4 Euthanasia, radiograph imaging and sample harvest  45  4.4 Sample analysis  45  4.4.1 Histology sample preparation  45  4.4.2 Histology and histomorphometry analysis  47  C H A P T E R 5: R E S U L T S  51  5.1 The effects o f coating parameters on calcium phosphate coating morphology  51  5.2 Calcium phosphate coating and alendronate loading  54  5.2.1 E L D coating o f calcium phosphate  54  5.2.2 The alendronate loading amount and estimation o f dosage  56  5.3 Crystal structure o f E L D calcium phosphate coating  56  5.4 Histology  59  v  5.5 Histomorphometry  61  5.5.1 Bone formation  61  5.5.2 Bone/implant contact length  64  5.6 Bone formation mechanism  66  C H A P T E R 6: D I S C U S S I O N  73  6.1 The effects o f calcium phosphate coating  73  6.2 The effects o f alendronate-immobilized calcium phosphate coatings  75  6.3 Clinical relevance o f the gap model animal tests and alendronate delivery implants ...77 6.4 K e y progresses in E L D surface drug delivery and animal studies  79  C H A P T E R 7: C O N C L U S I O N S  81  C H A P T E R 8: R E C O M M E N D A T I O N F O R F U T U R E W O R K  83  REFERENCES  84  APPENDICES  92  Appendix A : Student's t table (reprinted from ref 82 courtesy o f Deacon J.)  92  Appendix B : Ethical approval  94  vi  LIST OF TABLES Table 1.1  Total numbers o f hip and knee replacement procedures, Canada [3]  Table 4.1  The weight o f porous tantalum implant plugs  32  Table 4.2  A n i m a l test plan for bisphosphonate-coated porous T a bone implants  41  Table 4.3  Information o f fluorochomes and their solutions  44  Table 4.4  Labeling administration dosages and schedule  44  Table 4.5  Processes and schedules for fixation, dehydration and infiltration  45  Table 5.1  The statistic results o f bone formation  64  Table 5.2  The statistic result on contact length  65  vii  1  LIST OF FIGURES 5  Fig. 2.1  The illustration o f an artificial hip joint,  Fig. 2.2  The acetabular shell made o f porous tantalum (reprinted from ref. 14, with .5  permission from J o f Bone and Joint Surg.) Fig. 2.3  The 7 hierarchical levels o f organization o f the bone family o f materials. Level 1: Isolated crystals from human bone (left side) and part o f an unmineralized and unstained collagen fibril from turkey tendon observed in vitreous ice in the TEM(right side). Level 2: T E M micrograph o f a mineralized collagen fibril from turkey tendon. Level 3: T E M micrograph o f a thin section o f mineralized turkey tendon. Level 4: Four fibril array patterns o f organization found in the bone family o f materials. Level 5: S E M micrograph o f a single osteon from human bone. Level 6: Light micrograph of a fractured section through a fossilized (about 5500 years old) human femur. Level 7: Whole bovine bone (scale: 10 cm), (reprint from ref 16 with permission from Annual Review)  Fig. 2.4  8  Hierarchical structural organization o f bone: (a) cortical and cancellous bone; (b) osteons with Haversian systems; (c) lamellae; (d) collagen fiber assemblies o f collagen fibrils; (e) bone mineral crystals, collagen molecules, and non-collagenous proteins, (reprinted from ref 17 with permission from  9  Elsevier) Fig. 2.5  Logarithm o f the product o f calcium and phosphate concentrations plotted against p H values o f solutions saturated with respect to various calcium phosphate phases in the ternary system C a ( O H ) 2 - H P 0 - H 2 0 Calculated for 3  viii  4  37°C.(reprinted from ref 50 with permission from American Chemistry Society Publications) Fig. 2.6  18  S E M images o f biomimetic coatings o f CaP on porous tantalum (reprinted from ref. 56, with permission from Springer)  20  Fig. 2.7  S E M image o f the E L D coating o f calcium phosphate  22  Fig. 2.8  Chemical structure o f alendronate (reprinted from ref. 66, with permission from Springer)  23  Fig. 4.1  S E M images o f porous tantalum cylinder at various magnifications  32  Fig. 4.2  The image o f E L D coating cell  34  Fig. 4.3  Mechanical drawings o f porous tantalum implant assembly. The left is a 3D image o f the implant assembly, and the right is a 2-D drawing with all the dimensions  38  Fig. 4.4  The image o f bone cement casting mold  38  Fig. 4.5  Surgical tools and porous tantalum implant  39  Fig. 4.6  The images and schema o f the implant location, a) digital image o f implant position during surgery; X-ray image shows the implant position after surgery; c) the schematic image  43  Fig. 4.7  Schema for sectioning and observation directions  46  Fig. 4.9  Schema image for 3 grinding sections  46  Fig. 4.9  Histogram for image analysis and the images before and after dye process  49  Fig. 4.10  The measuring images for contact length and the available metal length  49  Fig. 4.11  The schema drawings illustrate the bone ingrowth and gap  50  ix  filling  Fig. 5.1  S E M images o f calcium phosphate coatings, a) without sulfuric acid cleaning, the porous tantalum strut was not coated; b) spallation o f calcium phosphate coatings; c) coatings morphologies at central pores without periodical stirring; (d) coating morphology at superficial layer when periodical stirring was not applied  Fig. 5.2  52  S E M images on calcium phosphate coatings at a central pore when periodical stirring was applied  Fig. 5.3  53  S E M images at varying magnifications o f Ca-P coating morphology, a) digital image o f implant assembly; b) low magnification S E M image shows the coverage o f CaP coating; c) and d) the porous morphology o f CaP coating  55  Fig. 5.4  S E M image showing coating thickness  55  Fig. 5.5  X R D results for CaP coatings before and after alendronate immobilization  57  Fig. 5.6  The  FTIR  results  for  CaP coatings  before  and  after  alendronate  immobilization Fig. 5.7  58  B S E and fluorescence images o f bone formations on 3 types o f implants, a) and b) non coated control; c) and d) calcium phosphate coated control; e) and f) alendronate immobilized calcium phosphate coated implant  Fig. 5.8  The mean and standard deviation o f volume o f gap filling, bone ingrowth and  total  bone  formation.  Ta-CaP-ALN  implants  demonstrated  significantly increased gap filling, bone ingrowth and total bone formation compared with Ta-CaP controls (p < 0.0005, p < 0.0001 and p < 0.0001  x  60  respectively *), and significantly increased gap filling, bone ingrowth and total bone formation compared with Ta control (p < 0.0001, p < 0.0001 and p< 0.0001 respectively **). N o significant differences in gap filling, bone ingrowth and total bone formation were found between Ta-CaP and Ta. (p = 0.12, p = 0.07, p = 0.07) Fig. 5.9  63  Bone formation on the surface o f porous tantalum expressed as a percentage o f the available porous tantalum surface (contact length) o f three  sections  performed  through  each  implant.  Ta-CaP-ALN  demonstrated significantly increased contact length compared with TaCaP controls, p < 0.0001 *, and significantly increased contact length compared with Ta controls, p < 0.0001 **. Ta-CaP controls demonstrated significantly increased contact length compared with T a controls, p < 0.02 + Fig. 5.10  66  Fluorescence images showing the labels on tooth and ingrowth bone, and the ordinal numbers gave postoperative time sequence, a) 4 color rings indicated the growth direction on tooth. The 2 green rings were calcein and 2 red rings were alizarin; b) labels indicated the bone initiation on implant surface  Fig. 5.11  67  Fluorescence images showing bone initiation, a) non coated porous tantalum control; b) alendronate immobilized calcium phosphate coated implant; c) and d)calcium phosphate coated controls  xi  69  Fig. 5.12  Fluorescence images showing bone ingrowth pattern on T a control, a) low magnification image; b) higher magnification image showing the labeling lines  70  Fig. 5.13  Fluorescence images showing bone ingrowth pattern on Ta-CaP control  70  Fig. 5.14  Fluorescence images showing bone ingrowth pattern on T a - C a P - A L N implant  Fig. 5.15  ...70  Fluorescence images showing the bone formation at gap areas. The white lines with double arrows indicated the bone bridging gaps a) T a control; b) Ta-CaP control; c) and d) T a - C a P - A L N implants  xii  71  LIST OF ABBREVIATION ALN  Alendronate  BMP  Bone morphogenic protein  BPP  Bisphosphonate  BSE  Backscattered Electron imaging  CaP  Calcium phosphate  DCPD  Dicalcium phosphate dihydrate  ECD  Electrochemical deposition  EDS  Energy dispersive spectrometry  ELD  Electrolytic deposition  FTIR  Fourier transform infrared spectroscopy  HA  Hydroxyapatite  HLPC  H i g h performance liquid chromatography  OCP  Octacalcium phosphate  PMMA  Polymethylmethacrylate  SD  Standard deviation  SEM  Scanning electron microscopy  Ta  Tantalum  Ta-CaP  Calcium-phosphate-coated porous tantalum  T a - C a P - A L N Alendronate-immobilized calcium-phosphate-coated porous tantalum TCP  Tricalcium phosphate  THR  Total hip replacement  XRD  X-ray diffraction  xiii  ACKNOWLEDGMENTS I would like to thank my supervisor, Dr. R i z h i Wang for his guidance during my research work, and Dr. D o n Garbuz for supervising surgical protocols during animal tests.  I also would like to thank my colleagues, K e Duan for his help throughout the entire project; Dr. Winston K i m for his excellent work on implant surgeries; Dr. T o m Oxland, Bas Masri and Clive Duncan for support in surgery designs; and A l l e n Tang for his support in biomaterial laboratory and in the Jack B e l l Centre o f the Vancouver General Hospital.  I would like to extend my thanks to Dr. Helen Burt and M r . John Jackson in the Pharmaceutical  Science Department o f U B C ; M s . Mary Mager and other staff in the  Materials Engineering Department o f U B C and the animal care team at Jack B e l l Centre in VGH.  Finally, Zimmer Inc. and N S E R C , thank you for your support with the  xiv  project.  CHAPTER 1: INTRODUCTION Osteoporosis and osteoarthritis may cause severe damage to the hip and/or knee joints. Injuries occurring during sports and exercises may also break the femoral necks or the knee patellae. To treat these diseases and traumas, orthopaedists often have to replace the failed hip or knee joints with artificial ones made o f metals, polymers or ceramics. Women are more vulnerable to osteoporosis [1]. After menopause a woman can lose up to 20% bone mass i n 5 to 10 years [2], and this may cause brittleness o f bone.  In the year 2004 - 2005, statistics showed that the number o f knee replacements in Canada reached 33.6 thousand, which was an increase o f 125% from the year 1994-1995, Tab 1.1. A l s o in 2005, hip replacements in Canada reached 25 thousand, an increase o f 52% and 6% compared with years 1994 -1995 and 2003 - 2004 respectively [3]. Considering that the population o f Canada in 2005 was 30 millions, almost two in every thousand Canadians have had either a hip or knee replacement in every year for the past 10 years. Therefore, any improvement in knee or hip replacements w i l l significantly contribute to the Canadian health care system and dramatically benefit Canadian society.  Table 1. 1  Total numbers o f hip and knee replacement procedures, Canada [3]  Knee Replacements H i p Replacements Total Replacements  1994-1995  2003-2004  2004-2005  14,938 16,525 31,463  29,848 23,669 53,517  33,590 25,124 58,714  1  10-year % Change 125% 52% 87%  1 - Year % Change 13% 6% 10%  Currently, poor fixation and unacceptably high revision rates are the major problems with knee and hip joint replacements. Poor early bone/implant integration, stress shielding and aseptic loosening are the major causes o f failure of joint replacement implants [4, 5], and the longevity o f these joint implants is 15 to 20 years. With the extension o f human lifespan to 80 or 85 years in the 2 1 century, a revision joint arthroplasty may be needed. Therefore, st  enhancing the bone/implant fixation and reducing the revision rates o f joint replacements remain the primary topics in orthopaedic research. In addition, the management o f bone stock deficiency and the achievement o f reliable and durable implant/host-bone fixation remain significant challenges in revision joint replacements.  The concept o f a drug-implant delivery system in reconstructive orthopaedics was well established [6]. Bisphosphonates have been investigated as a means to promote early bone ingrowth [ 7 , 8, 9 ] .  Recently, an investigation o f a canine model where a third  generation bisphosphonate, zoledronic acid, bound to hydroxyapatite-coated porous tantalum implants demonstrated an increase in peri-implant bone growth within the intramedullary canal and a 58% increase in the bone ingrowth within the porous tantalum[10].  Laboratory experimental models should replicate clinical scenarios as closely as possible. In revision hip arthroplasty, host bone deficiencies, gaps and defects between implant and host bone are frequently encountered. Gaps as small as 50 to 500pm failed to be bridged by new bone in a study o f human host bone and hydroxyapatite-coated implants [11]. The design o f implant materials has to face challenges o f the inevitable gaps between bone  2  and implants during arthroplasty surgeries, and the gap model animal tests are necessary for examining new designed implant materials.  Although the promising results on bisphosphonate-delivery implants were reported [10], the effects o f bisphosphonate-immobilized calcium-phosphate coating on the location, pattern and extent o f bone formation in the presence o f a gap have not been previously investigated. We conducted a gap model animal study on alendronate-immobilized calciumphosphate-coated porous tantalum implants. The calcium phosphate coatings we processed via electrolytic deposition were porous at 1 micrometer level, and the porous morphology was good for bisphosphonate immobilization. L o w concentration (less than 10" mol/1) o f 4  alendronate, which was one o f the bisphosphonate families, was immobilized on the surfaces o f E L D calcium phosphate coating. W e hypothesized that the alendronate-immobilized calcium phosphate coating could increase the osteoconductivity and bone ingrowth o f porous tantalum implants and therefore enhance the peri-prosthetic bone growth. Eventually, better bone/implant fixation could be achieved.  The purpose o f this study was (1) to develop an experimental animal model which would simulate the clinical revision hip scenario, (2) to determine the effects o f alendronateimmobilized calcium phosphate coating on porous tantalum on gap filling and bone ingrowth in a rabbit model, and (3) to develop techniques o f improving the bone/implant fixation o f porous tantalum joint replacements.  3  CHAPTER 2: LITERATURE REVIEW 2.1 Introduction to artificial hip joints Since the 1960s, when the artificial hip joint with polymer liner was developed and applied by John Charnley [12], many improvements have been made, and the total hip replacement ( T H R ) has become an industry worth billions o f dollars. Nowadays, many different materials such as cobalt-chromium alloys, titanium alloys and tantalum are being used to ensure the biocompatibility o f T H R , and different rough surface treatments such as beads and meshes are employed to enhance bone attachment to T H R implants. The biocompatibility here is defined as "the ability o f a synthetic or natural material for medical implantation to receive the appropriate reaction from recipient environment" [13].  A n artificial hip joint typically includes an acetabular shell for attaching to the pelvis, a polyethylene liner for reducing friction and an artificial femoral head and stem for inserting into the femoral intramedullary canal. One example o f these hip joint replacements is a titanium alloy stem with dents o f 2 to 3 millimeters and an acetabular shell with porous titanium alloy fiber mesh on top(Fig. 2.1).  Recently, porous tantalum,  also known as trabecular  metal, was adopted  for  manufacturing the acetabular shell (see F i g . 2.2) [14]. This sponge-like porous tantalum had 400 to 500 micrometer pores that allow bone to grow into them. The outcome o f this implant design was the mechanical interlock o f the pelvis bone with the acetabular shell, resulting in improved bone / implant fixation.  4  Fig. 2.1 The illustration o f an artificial hip joint  Fig. 2.2 The acetabular shell made o f porous tantalum (reprinted from ref. 14, with permission from J o f Bone and Joint Surg.)  2 . 2 Bone Knowledge o f bone, such as its physical and chemical properties, structures, and reactions to wounds etc., is essential to developing joint replacements. Bone is the main in vivo environment that joint replacements deal with, and bone is also the target that implants should imitate. A l s o , bone w i l l undergo a healing process due to the interference caused by implantation. A brief review is provided to illustrate the structure, mechanical properties, and the formation and healing process o f bone.  2.2.1 Structure of bone The structural components o f bones are calcium minerals, collagen proteins, water, and some non-collagen proteins. The two main components o f bone are minerals and collagen proteins, which contribute to more than 90 % o f its weight [15,16]. The bone minerals are primarily carbonated apatite plates with dimensions o f up to 50 nm by 25 nm by 3 nm. The collagen proteins are mainly type I and fibrous, and assembled from triple-helical collagen molecules that are 300 nm in length by 1.2 nm in diameter (see F i g 2.3 Level 1 and F i g 2.4 Sub-nanostructure). The mineral plates and collagen fibers assemble together to form the mineralized collagen fibrils [17].  The mineralized collagen fibrils are the building blocks o f bones (see F i g 2.3 Level 2 and F i g 2.4 Sub-nanostructure). These mineralized fibrils are 50-100 nm in diameter and up to a few millimeters in length. Bundles o f collagen fibrils form individual sheets, or bone lamellae, which stack together to make the lamellar structure [16].  The Haversian system, or the secondary osteon (Fig 2.3 Level 5 and F i g 2.4 Microstructrure), is a small unit o f bone with central blood canals (Haversian canals, about 50 um in size) surrounded by bone lamellae. When bone becomes old or fractured, bone cells start the remodeling process [18, 19]. Osteoclasts resorb the bone and generate the tunnels. Following osteoclasts, osteoblasts rebuild the bone layer by layer as lamellar bones. The central tunnel also functions as a blood vessel with smaller channels o f canaliculi connected to it. The canaliculi also connect to osteocytes and supply them with nutrition.  A t the macroscale level (Fig 2.3 Level 6 and F i g 2.4 Macrostructure), there are two types o f bones classified on the basis o f porosity and unit microstructure. Cortical bone, also known as compact bone, is much denser with a porosity o f 5 and 10%. It consists o f Haversian systems and has fewer blood vessels and bone cells inside it. A l s o , cortical bone has higher mechanical properties (Young's modulus: less than 30GPa). In comparison, spongy bone, also known as 'cancellous bone' or 'trabecular bone', has a porosity o f 50% to 90%). Trabecular bone has much more blood vessels and bone cells, but its mechanical properties such as strength and modulus are lower (Young's modulus o f architectural trabecular bone: 0.8GPa).  The pelvis bone is mainly made o f the trabecular bone. For a long bone such as a femur, the shaft is mainly made o f cortical bone surrounding the marrow. Trabecular bone is found at the two ends (distal and proximal) o f the femur next to the joints. The pelvis and the femur are the places where T H R s are located.  7  Level 1 : Major Components  Fig. 2.3 The 7 hierarchical levels o f organization o f the bone family o f materials. Level 1: Isolated crystals from human bone (left side) and part o f an unmineralized and unstained collagen fibril from turkey tendon observed in vitreous ice in the TEM(right side). Level 2: T E M micrograph o f a mineralized collagen fibril from turkey tendon. Level 3: T E M micrograph o f a thin section o f mineralized turkey tendon. Level 4: Four fibril array patterns o f organization found i n the bone family o f materials. Level 5: S E M micrograph o f a single osteon from human bone. Level 6: Light micrograph o f a fractured section through a fossilized (about 5500 years old) human femur. Level 7: Whole bovine bone (scale: 10 cm), (reprint from ref 16 with permission from Annual Review o f Materials Science)  8  Collagen molecule  L  Cancellous bone Collagen fiber  Lamella Cortical bone Osteon  Bone Crystals  Haversian canal  10-500 Jim  Microstructure Macrostructure  Nanostructure  Sub-microstructure  Sub-nanostructure  Fig. 2.4 Hierarchical structural organization o f bone: (a) cortical and cancellous bone; (b) osteons with Haversian systems; (c) lamellae; (d) collagen fiber assemblies o f collagen fibrils; (e) bone mineral crystals, collagen molecules, and non-collagenous proteins, (reprinted from ref 17 with permission from Elsevier)  2.2.3 The mechanical properties of bone The mechanical properties [20] o f lamellar bone are anisotropic. A l o n g the long bone axis the mechanical properties such as strength and modulus etc. are higher than those transverse to long bong axis. With the tested angles varying from 0 to 90 degree, the mechanical properties reduce by 3 to 5 times. For example, the bending strengths reduce from 3 5 0 M P a or higher at 0 degree to lOOMPa at 90 degree; flexural modulus decrease from 16GPa to 7GPa; and the work o f fracture (the energy to break the material) reduces from 8kJ/m to less than l k J / m [21]. The reason for these differences is because o f the fibril 2  2  structure inside lamellar bone, in which collagen fibrils are parallel to the long bone axis. A l s o , the body weight applies loadings on the long bone axis, and so they are expected to be stronger in that direction.  9  2.2.4 Bone formation and bone healing process Bone healing is a complex biological process [22]. It needs the coordinated actions o f growth factors, osteoclasts, osteoblasts and immune cells, etc. within the bone marrow. Stem cells that are recruited from the surrounding tissues and the circulation w i l l differentiate into osteogenic cells. Multiple factors control these chains o f reactions and further affect the different  sites in the osteoblast  lineage through various processes such as migration,  proliferation, differentiation, inhibition, and protein synthesis.  Cortical bone and trabecular bone experience different healing processes [ 2 3 ] . Trabecular healing depends on osteoconduction and de novo bone formation. O n the other hand, cortical healing relies on the remodeling o f bone osteons.  The de novo bone formation can be divided into four-steps [24]. First, differentiating osteogenic cells, which are the cells still possessing traveling capacity but w i l l become osteoblasts, produce collagen-free organic matrix that provides nucleation sites for calcium phosphate mineralization. Then, calcium phosphate nucleation happens on the collagen-free organic matrix, and the initiation o f collagen fiber assembly follows as the third step. Finally, calcification o f the collagen compartment w i l l occur.  Cortical  healing process  can  be  divided into  five  major  periods  including  inflammatory, granulation tissue formation, callus formation, lamellar bone deposition and remodeling [25]. A n d osteoclasts involve in bone healing at two different periods. The population and activation o f osteoclasts are extremely high at the early stage o f bone healing  10  because the major function o f osteoclasts is to resorb bones. They clean up the broken bone and supply minerals and proteins for the formation o f new bones.  The second period we see the high activation o f osteoclasts is in the bone remodeling process [25]. The newly formed bone during fracture healing is woven bone, which does not have the regular bone structure. The woven bone can be regenerated fast and bear certain loads. This helps to quicken the healing o f injuries. The woven bone w i l l be remodeled later to the regular structure with osteons, canaliculi and nerves, etc. A t the remodeling stage, osteoclasts resorb the woven bone and create tunnels, and then the osteoblasts regenerate the lamellar bone surrounding those tunnels. The combination o f osteoclasts and osteoblasts activities generates osteons and completes the bone healing process.  Contact osteogenesis and distance osteogenesis [23, 26] are the.criteria used to determine whether implant surfaces are osteoconductive (defined on page 12). Contact osteogenesis means that the bone formation happens on the surface o f or around the implants, and distance osteogenesis is when the bone starts growing at the site o f host bone. If the implant surfaces are osteoconductive, they should induce the bone formation towards them, and both contact and distance osteogenesis w i l l occur. Otherwise, the contact osteogenesis w i l l not happen, and only distance osteogenesis w i l l take place.  2.3 Implant materials Material selection is very important in artificial joint replacements. The physical and chemical properties o f the chosen materials have to fit in the applied physiological  11  environment and meet the mechanical requirements. Corrosion o f implants in an in vivo environment is a great concern [27]. O n the another hand, the implanted materials must establish a firm interface with the adjacent tissues such as bone, and satisfy the loading needs of natural bone.  The materials suitable for artificial joints include several metals or metal alloys such as stainless steels, titanium, titanium alloys, cobalt-chromium alloy, and tantalum. A s these metals or alloys are chosen mainly to fit the biomechanical requirements, the interaction with biological environment is compromised. These metallic implants are usually bioinert and have  very  limited  interactions  with surrounding  bone  tissues.  Achieving  a  strong  implant/bone interface has been a challenge [28, 29].  Concepts such as osteoconductivity and osteoinductivity need to be clarified. Both osteoconductivity and osteoinductivity are indicators o f the capabilities o f bone formation onto bone grafts and implants. Osteoconductivity [30] defines the ability which the graft (porous implant as well) supports the attachment o f new osteoblasts and osteoprogenitor cells. It provides an interconnected structure through which new cells can migrate and new vessels can form. Osteoconductivity is the ability o f bone graft (or porous implant) to enable new bone to grow through it. Osteoinductivity [30] refers jto the ability o f a graft (or porous implant) to induce non-differentiated stem cells or osteoprogenitor cells to differentiate into osteoblasts.  12  Tantalum is one o f the metallic materials that have excellent corrosion resistance [31]. It hardly stimulates adverse biological response. Many studies have demonstrated  the  excellent biocompatibility o f tantalum as in bone implants [31, 32, 33]. Solid tantalum may not be good for bone implants because o f its high density (16.7 grams/cm ). The implants 3  made o f solid tantalum are much heavier than bone they replace and bring much higher loads to the bone skeleton. The high modulus (200GPa) o f solid Tantalum (in comparison to modulus o f cortical bone: less than 30GPa) may also cause a 'stress shielding' problem (see page 15).  Porous  Tantalum is  an  ideal implant material  that  combines  biocompatibility with its mechanical properties similar to those o f bone.  its  excellent  The clinically  available porous tantalum has 75% to 80% porosity, and its elastic modulus is 3 G P a [34]. "The compressive and shear strengths o f this porous tantalum are 35 to 40 M P a . " [31, 34] Another advantage o f porous tantalum is that bone can grow into porous tantalum and construct an interlock system between bone and porous tantalum implants, and the interlock between bone and implant resulted in better bone/implant fixation [34, 35].  The manufacture o f porous tantalum is done through a chemical vapor deposition process. Obtaining the desired porosity in the carbon skeleton is achieved through pyrolysis o f polyurethane sponge. The tantalum is deposited on the carbon skeleton by chemical vapor deposition with a tantalum pentachloride precursor [34].  13  2.4 The challenges in hip replacement Poor implant/bone fixation and an unacceptably high revision rate are the major problems with current joint replacements [4, 5]. Aseptic loosening [36], which is caused by periprosthetic bone resorption (also called osteolysis) triggered by the polyethylene wearing debris entering the joint space [37], is one o f the main causes o f the implant failure. Another reason is 'stress shielding' which is caused by the metallic materials used for femoral stems.  The present implantation technology is adequate for pain relieving and movement restoring to the ailing hip joints, and the hospitalization time is usually 6 days and full recovery takes about 4 to 6 months [38]. However, many challenges remain unsolved. A s the lifespans o f human beings continue to increase i n the 2 1 century, the number o f patients st  who need T H R w i l l also increase because o f the bone problems such as osteoporosis and osteoarthrisis. In 2001, the waiting time for T H R surgery in Canada was from 6 to 23 months depending on which province patients lived in [39]. Patients have to endure the serious pain or sit in wheelchairs waiting for their arthroplastic surgeries.  A t the same time, the designed implants should extend their longevity to meet the requirement o f long post-surgery living time. The existing hip replacements normally can serve patients up to 15-20 years and 20 percents o f these hip replacements fail within 20 years. This may cause serious pain to the patients, and new revision surgeries w i l l be needed.  The  biological processes o f periprosthetic bone resporption (leading to aseptic  loosening) include phagocytosis on the wear particles by macrophages, ingestion o f debris  14  particulates and eventually the bone resorption [37, 40]. When wear particles are present, the instinctive reaction (called phagocytosis) o f the immune system is to deploy the macrophages to consume these alien particles. Then, the ingestion o f particulates o f debris starts, and the cytokines and other mediators-of inflammation are released simultaneously. These factors then lead cytokines, and other biological components, to activate osteoclasts to resorb bone around the bone/implant interface. Other than polyethylene particulates, metal particles and bone cement particulates also contribute to osteolysis in the same way. The wear debris associate with periprosthetic osteolysis poses a long-term threat to implant longevity [41].  Stress shielding [42] is associated with the load sensing capability o f bone and poses a challenge to hip replacements. The lower limbs sustain the load from the body weight and the muscles attached to the long bones exert stresses on the skeleton. Bone is a 'smart material' i n that it grows more bone for supporting where the load on bone is larger. For the same reason, it w i l l grow less bone where the load on bone is smaller. The femoral stem o f a hip implant (for example, the Young's modulus o f T i alloy is 115GPa) is much stiffer than the cortical bone (Young's modulus is less than 30 GPa) and w i l l take the greater part o f the load due to body weight. A s a consequence, the femoral bone around the stem is "unloaded", and the femoral bone at the tip o f femoral stem is heavily loaded. The loading changes by introducing joint implants to the skeleton induce bone density changes throughout the femur. The  upper part o f the femoral bone w i l l contain less bone tissue and become more  susceptible to fracture. The bone at the tip o f implant stem becomes thicker and stronger. Unfortunately, the thickening o f the skeleton is often painful [ 4 3 ] . The patients with  15  cementless stems o f total hip devices often feel pain in the thigh, especially during the first year after the surgery.  A major challenge in revision hip arthroplasty is achieving early and durable implanthost bone fixation, especially in the presence o f significant bone stock deficiencies. Traditional uncemented implants (porous -coated and titanium fiber mesh) have shown good results clinically in the revision setting. However i n cases where less than 50% host bone contact is made the results o f these uncemented implants has been less predictable [44].  The message from orthopaedists is that these bone deficiencies, bone/implant gaps and  defects  are  inevitable during the  arthoplastic  surgeries.  Autopsy studies  have  demonstrated that bone ingrowth occurs in only 30-40% o f the surface area o f the implant in primary total hip arthroplasty [31, 45]. In revision arthroplasties, bone defects and gaps may significantly compromise the quantity and quality o f host bone available for ingrowth. In many cases less than 50% o f the host bone is available for ingrowth o f the newly implanted shell. To any porous implant material, the reliable and durable bone ingrowth with limited host bone contact is considered a desirable improvement. The improvement would result in an implant that is more versatile in the challenges o f revision cases.  2.5 Current material approaches to improve implant-bone fixation To achieve early implant-bone fixation, many new approaches have been launched to enhance the bone growth on the metal implant surface. Hydroxyapatite, bone morphogenic proteins ( B M P ) and other materials are introduced either for their similarities o f crystal  16  structure to bone minerals or for their biological functionalities in the bone formation processes.  Sumner et al [ 4 6 , 47] have studied the effects o f B M P on post-surgical bone formation in a canine model. They dipped the implants into different types o f B M P s , such as r h B M P - 2 , before the implant surgeries and examined the bone growth 4 weeks after surgery. The results were positive on enhancements o f B M P s on bone formation. Recently, they conducted new animal tests o f a gap model with combined addition o f B M P - 2 and the transforming growth factor P 2 (rhTGF-(32) on hydroxyapatite-plasma-sprayed  titanium  implants [48]. Compared with non-treated control samples, significant increases in bone formation towards implants were reported on B M P s treated implants. The downsides o f B M P treated implants are their low efficiency due to the quick liberation o f B M P , the unexpected calcification o f surrounding soft tissue, the challenges o f obtaining growth factors, storage and transportation, and their low versatility.  2.6 Calcium phosphate There are several forms [49] o f calcium phosphate. Dicalcium phosphate dihydrate ( D C P D ; C a H P 0 - 2 H 0 ) , tricalcium phosphate ( T C P ; C a ( P 0 ) 2 ) , octacalcium phosphate 4  2  4  3  ( O C P ; C a H 2 ( P 0 4 ) 6 - 5 H 0 ) and hydroxyapatite ( H A ; C a , o ( P 0 ) 6 ( O H ) ) are commonly used 8  for  4  2  implants or implant coatings.  physiological  condition  is  2  The solubility o f these calcium phosphates in a  generally  in  following  order  from  highest  to  lowest:  D C P D > O C P > T C P > H A . The p H o f the solution also affected the solubility o f these calcium  17  phosphates. In general, as the p H increases from 3 to 9, the solubility o f all four forms o f calcium phosphates decreases [50, 51].  Because o f the similarity o f hydroxyapatite to bone mineral contents, hydroxyapatite has  been  studied as an implant material for many  years.  The crystal structure  of  hydroxyapatite is hexagonal with a=9.4 A and c=6.9 A cell dimensions [52]. A s a type o f ceramic, hydroxyapatite is considered brittle. Its fracture toughness (KJC) is less than 0.9 MPa-m  , and compressive and tensile strengths are less than 300 M P a and 50 M P a  respectively. Based on its l o w mechanical properties, hydroxyapatite alone can not be the load bearing prosthetic material for bone. It is often used as a coating on metallic implants.  30  40  50  ,60 pH  70  8-0  90  F i g . 2.5 Logarithm o f the product o f calcium and phosphate concentrations plotted against p H values o f solutions saturated with respect to.various calcium phosphate phases in the ternary system C a ( O H ) 2 - H P 0 4 - H 2 0 Calculated for 37°C.(reprinted from ref 50 with permission from American Chemistry Society Publications). 3  18  Phosphate conjugates into several forms such as PO4 ", HPO4 " and H2PO4". In diluted 3  2  aqueous solution, the existences o f phosphate forms depend on the p H o f solution. When the solution is strongly basic (high pH), the phosphate mainly exists as phosphate ions (PO4 "); when the solution is mildly basic, the phosphate is mostly hydrogen phosphate ions ( H P C V - ) ; when the solution is mildly acidic, the phosphate is comprised o f dihydrogen phosphate ions (H2PO4"). This knowledge is essential for the deposition o f calcium phosphates from an aqueous solution.  2.7 Deposition of hydroxyapatite There are many methods for depositing hydoxyapatite on metal surfaces. These methods include plasma spray, biomimetic deposition, sol-gel deposition and electrolytic deposition and so on. Electrolytic deposition ( E L D ) , which is also called electro chemical deposition ( E C D ) , is one o f the non line-of-sight coating methods that are suitable for depositing hydroxyapatite on complex 3-D surfaces. Therefore, E L D is the ideal method for depositing hydroxyapatite on porous tantalum.  Plasma spray deposition o f hydroxyapatite[ 53 ] is the most popular method for orthopaedic implants. The hydroxyapatite powder is heated to a very high temperature (30,000K [54]) by plasma and sprayed on implant substrates. The coatings are lamellar structure [55]. A l s o , the melted hydroxyapatite may change to other forms o f calcium phosphates such as tricalcium phosphate or even amorphous calcium phosphate [55]. There is hardly any chemical bonding between the substrate and hydroxyapatite coating. Thermal stresses are introduced to coatings and to the coating/substrate  19  interface by the high  temperature used in heating. Plasma spray is a line-of-sight process which is not suitable for complex 3 D structures such as porous tantalum because only the superficial pores can be coated.  Biomimetic coating o f calcium phosphate [56] is a process that mimics the natural mineralization mechanisms in bone. During the coating process, substrates are simply immersed in a supersaturated calcification solution and the coatings grow onto substrates in matter o f hours to days. Biomimetic coating o f calcium phosphate is attractive because the obtained crystals bear more similarities to the composition o f bone minerals, its morphology and structure, and the heat treatment o f coating can be eliminated. A l s o , this method is non line-of-sight and is suitable for complex structures. C u i et al [56] coated porous tantalum using the biomimetic method and obtained a flower-like calcium phosphate coating o f calcium phosphate (see Fig.2.2).  Fig. 2.6 S E M images o f biomimetic coatings o f CaP on porous tantalum (reprinted from ref. 56, with permission from Springer).  The downside o f the biomimetic method is the requirement o f a strict chemical environment and surface treatment. For the H A crystals to nucleate, the surfaces o f prosthetic  20  metals need to be chemically etched (e.g. acid/alkaline treatment) or functionalized with negatively charged groups such as sulfonate [57], carboxyl and phosphate [58]. O n the native T i surface without chemical modification, as crystallization theory predicts, the synthesizing parameters need to be strictly controlled [59].  Electrolytic deposition o f calcium phosphate is another non line-of-sight process. It takes advantage o f the electrode reactions in aqueous solution, and deposits the calcium phosphates on the cathode. When the electrical field is applied, the following cathode reactions w i l l happen [60]: 2 H + 2e -> H t +  2  2 H 0 + 2e -> 2 0 H " + H T 2  0  2  2  (aq) + 2 H 0 + 4e-+ 4 OFT 2  These reactions w i l l raise the p H o f the solution surrounding the cathode. A s mentioned above, the solubility o f calcium phosphates ( D C P D , O C P , T C P and H A ) drops while the p H of the solution increases (see page 17 and 18). B y controlling the p H o f electrolyte and the concentrations o f calcium and phosphate ions, different forms o f calcium phosphates may deposit on the cathode where the prosthetic metal substrates locate.  Our group [61] has developed an E L D coating technique that can achieve porous structural coatings o f hydroxyapatite on titanium and tantalum. The porous structures o f hydroxyapatite with pore size o f 1 micrometer provide surfaces immobilization.  The combination o f the  E L D coatings  21  for bisphosphonate  o f hydroxyapatite and  the  immobilization o f bisphosphonates is one o f the promising surface treatments for porous tantalum orthopaedic implants.  Fig. 2.7 S E M image o f the E L D coating o f calcium phosphate.  2.7 Effect of bisphosphonates on bone Bisphosphonates are a family o f drugs that have been clinically used to treat osteoporosis and to prevent bone fracture [62, 63]. The effects o f bisphosphonates on bone are the decrease in the activities o f osteoclasts and the increase o f the activities o f osteoblasts. Both osteoclasts and osteoblasts, which resorb bone or deposit minerals on tissue respectively, are bone-related cells. The two effects o f bisphosphonates mentioned above, are the enhancement o f bone formation and an accelerated healing process after arthroplasty surgery [ 6 4 , 6 5 ] . Fig.2.8 is the molecular structure o f alendronate which is one o f the bisphosphonate families.  22  CH —NHj 2  CH  2  CH  2  HgOaP-C-POgHj OH Fig. 2.8 Chemical structure o f alendronate (reprinted from ref. 66, with permission from The American Society for Pharmacology and Experimental Therapeutics)  Another chemical characteristic o f bisphosphonates is their affinity with calcium ions. Bisphosphonates can chelate with calcium and form bisphosphonate and calcium bonds [59]. This characteristic enables bisphosphonates to be immobilized on hydoxyapatite and create an in vivo device that slowly releases bisphosphonates.  2.8 Current studies on bisphosphonate-delivery implants The effect o f zoledronate, which is one type o f bisphosphonates, on bone ingrowth into porous tantalum implants was examined in a canine model [67]. Zoledronate in saline was administered in a single post-operative intravenous injection at a dose o f O.lmg/kg. A t 6 weeks post-surgery, bone ingrowth increased by 85 % compared with controls. Subsequently, an investigation where zoledronate bound to hydroxyapatite coated (by plasma spray) porous tantalum was performed by the same group [68]. The study demonstrated an increase in periimplant bone within the intramedullary (inside bone marrow) canal (32.2 % versus 13.8 % in controls) and intra-implant bone formation (19.8 % versus 12.5 %). However, the authors acknowledged the effect o f the normal reparative stimulus to reaming the canal, resulting in greater peri-implant than  intra-implant bone  23  formation. A l s o ,  the  plasma spray o f  hydroxyapatite is a line-of-sight process in which only the superficial layer o f porous tantalum was coated.  Some researchers designed bisphosphonate-delivery methods on different implants materials targeting different implantation locations, and they reported positive effects on enhancing bone growth. Peter et al [69] soaked plasma spray hydroxyapatite coated titanium alloy in zoledronate for T H R , and Binderman et al [70] applied alendronate on alveolar surfaces before they implanted the dental replacements.  2.9 Gap model animal tests Traditional uncemented implants (porous coated and titanium fiber mesh) have shown good results clinically in the revision setting [71]. However in cases where less than 50% host bone contact is made the results o f these uncemented implants have been less predictable [44]. Porous tantalum is a relatively new porous biomaterial, which may represent an improvement from conventional uncemented materials. Its properties o f a high coefficient o f friction (average 0.86±0.11 and 0.98±0.17 against surface-flattened cortical bone and cancellous bone respectively; average 0.58±0.06 when cancellous bone against surface-flattened cortical bone [72]), interconnected pores and high porosity by volume (80%) [31, 45] make it an ideal material in the revision setting with compromised host bone.  The experimental animal tests should replicate the real case scenario during the arthroplastic surgeries. Orthopaedists cannot prepare a patient's pelvis to exactly match the artificial acetabular cup during arthroplastic surgery, especially in the case o f bone  24  deficiencies caused by trauma, bone cancer and former damages revealed in revision. The gaps between bone and the acetabular cup are inevitable. A newly developed implant not only has to enhance the grown bone-implant attachment, but also has to overcome the gaps between host bone and implant.  To simulate the real case, a gap model animal test is necessary. Studies [48,73] showed that gaps o f 0.5 to 3 mm inhibited bone ingrowth, and the inhibition increased with the size o f the gap. The gap model animal tests, should consider the bone size o f the tested animal and the implant size to decide upon the proper gap size.  2.10 Mechanism of labeling The fluorochromes can chelate with the calcium atoms in hydroxyapatite and attach themselves to newly mineralizing front o f bone [74]. Only the newly mineralizing surfaces can be labeled probably because o f the difference o f crystal sizes o f hydroxyapatite between initial stages o f mineralization and older mineralization sites [75]. The crystal size o f hydroxyapatite is smaller at initial stages o f mineralization than at older sites.  The  commonly used  fluorochromes  for  bone  labeling are  calcein, alizarin  complexone and tetracycline etc. [74]. O n retrieved bone samples, these fluorochromes reveal different colors under a fluorescent microscope with the proper light sources and filters. A t different postoperative time-points, different fluorochromes can be administrated, and the mineralization process o f newly formed bone is completely monitored [76]. However,  25  the injection dosages have to be carefully controlled because of the potential poisoning of these fluorescent chemicals.  Pautke et al [77] developed a practical labeling protocol on mice bones. They adopted 8 different fluorochromes and sequentially injected them into mice bodies with a 3 day interval. Eventually, they labeled 7 time-points during 4 weeks of bone growth. They claimed that they could observe 6 different fluorescent bands by means of spectral analysis after 6 months, and 4 bands by the naked eye with proper fdters.  Frosch et al [78] applied 4 different types of fluorochromes, which are xylenol, calcein, alizarin complexone and tetracycline, to study the enhancement of autologous osteoblast on ingrowth inside drilled 400-600 micrometer holes on titanium substrates. Specimen rabbits were injected with fluorochrome every 5 days postoperatively. The color labeling provided the 'wood pattern' rings of different post surgical time.  2.11. t-tests Statistics, as a mathematic tool, is often used for medical and clinical studies [79]. Researchers and medical doctors constantly need to draw conclusions on the effects of a pharmaceutical product or a medical procedure based on the symptoms of a small number of patients to those of the whole patient population. Individual patients differ from one another, and so it is essential to use proper methods in evaluating received data. This is the sample representation problem [79]. Another problem is the comparison among the data groups [79]. Based on the measured data, are there any differences among the data groups and are these  26  differences significant? t-test is one o f the statistical methods comparing two groups o f data at a time to answer these questions.  There are several different types o f t-tests, such as equal size or unequal size, dependent or independent, based on the two groups o f data [80]. The t-test is equal or unequal size depending on whether the two compared groups have equal data points or not. The t-test is dependent or independent regarding whether there is any relation between two compared groups or not. If one sample is tested twice or two samples are matched or 'paired', the t-test is dependent; otherwise it is independent.  Before t-test, we assume that the data points follow the 'normal distribution', and the means and the data distribution curves o f two compared groups are not different (null hypothesis) [80]. The null hypothesis is either sustained or rejected based on the t-test result and the selective confidence level. The confidence level is the degree o f certainty that a statistical prediction is accurate, and it is normally chosen to be 95 or 99%. p-value is "a measure o f probability that a difference between groups during an experiment happened by chance.... The lower the p-value, the more likely it is that the difference between groups was caused by treatment."[81] For instance, i f the confidence level is chosen to be 95%, p-value, as the result oft-test, has to be bigger than 0.05 (1 minus the confidence level 0.95) to sustain the null hypothesis. Otherwise the null hypothesis is rejected, and the difference between two compared groups is caused by treatment.  27  t-test starts with the calculations o f the arithmetic averages (X ) and the standard deviation (a) o f the data points o f the two compared groups (X 1 and o\ for group 1; X a  2  2  and  for group 2). N represents the number o f data in a group(Ni for groupl and N for group 2  2). The degree o f freedom is "a measure o f the number o f independent pieces o f information on which the precision o f a parameter estimate is based" (courtsey o f Wikipedia). For the independent t-test with unequal sizes, the degree o f freedom is the number o f data minus 1. Therefore the degree o f freedom in total is N i + N - 2 . 2  following (assume the X  x  is bigger than X  2  The t-value can be calculated as  )[82]:  Once the t-value is calculated, it can be compared with the tabulated t-value at the degree o f freedom and chosen confidence level (tabulated probability p equals 1 minus confidence level) from Student's t table (see Appendix A ) [82]. If the calculated t is bigger than the tabulated t, the difference o f the two compared groups is significant. In this case, the calculated t can be compared with higher confidence level. Eventually, the p-value can be decided. If the calculate t is smaller than the tabulated t, p-value o f t-test is considered bigger than the tabulated probability p, and the difference between the 2 compared groups is not significant.  28  CHAPTER 3 : SCOPE AND OBJECTIVES The scope of this research is to design a proper gap model of animal tests and to conduct one cycle of animal testing to study the effects of a newly developed alendronatedelivery-coating treatment on the osteoconductivity and bone ingrowth of porous tantalum implants. The experiments include implant processing such as electrolytic deposition of calcium phosphate and alendronate immobilization. Animal tests include surgeries and postoperative animal care, data collection and analysis include histology, histomorphometry and statistic analysis. The project requires collaboration with the Faculty of Pharmaceutical Sciences, the Department of Orthopaedics and the animal centre at the Vancouver General Hospital.  The newly developed coating treatment takes advantage of the effects of alendronate on bone cells to reduce bone resorption and enhance bone formation, and combines with the electrolytic deposition calcium phosphate coating as a delivery vehicle. E L D coating of calcium phosphate is a non line-of-sight process suitable for complex 3 D structures such as porous tantalum, and the porous morphology of calcium phosphate coatings provides surfaces for alendronate adsorption. The final product is a combined device of orthopaedic implants with a local alendronate, delivery function, which is expected to improve the bone/implant fixation.  The design of animal tests has to simulate the real clinical scenario and to separate the effects of alendronate from any other potential factors. The gap model design is employed  29  because the bone/implant gaps are inevitable during revision arthroplasty surgeries. The calcium phosphate  coatings on porous tantalums  are used as second controls. The  fluorescence labeling method is employed to track the new bone formation and therefore also allows for studying the mechanism o f bone formation.  The objectives o f this research are: 1. To modify an existing technique o f electrolytic deposition o f calcium phosphate to be suitable for porous tantalum substrates. 2.  To immobilize an effective amount (dosage o f 2 5 0 x l O " M / L to 10" M / L per day) o f 9  4  alendronate onto the calcium-phosphate-coated porous tantalum implants. 3.  To design a proper gap model o f animal tests that can replicate the real case scenario o f arthroplasty and revision surgeries.  4.  To conduct one cycle o f animal tests to evaluate the effects o f alendronate delivery coating treatment on osteoconductivity and osteoinductivity o f porous tantalum implants.  5. To study the bone formation mechanisms on different surface treatments o f porous tantalum implants for the further understanding and improvement o f implant surface treatments.  30  CHAPTER 4: MATERIALS AND METHODS Bone deficiencies, bone/implant gaps and defects are inevitable during revision arthoplastic surgeries [44]. The experimental designs o f animal tests should replicate the clinical scenario. The intention o f the gap model animal tests was to investigate the bone reaction towards different surface treatments o f implant materials in the presence o f a gap between the implant and bone. A s a result, the better implant surfaces in terms o f overcoming gaps and enhancing bone/implant fixation could be determined.  This animal study o f alendronate-immobilized calcium-phosphate-coated  porous  tantalum implants included implant preparation, sterilization o f implants and tools, implant surgeries, post surgery care o f animals, sample harvest and sample analysis. The implant preparation  included pre-coating surface  cleaning, electrolytic deposition o f calcium  phosphate, alendronate immobilization, cap mold design and bone cement cap manufacture, and cap assembling. The sample harvest and sample analysis step included euthanasia, femur dissecting and sectioning, bone histological processing and embedding, and histology and histomorphometry analysis.  4.1 Implant preparation 4.1.1 Porous tantalum The porous tantalum implant (Trabecular Metal ™ ) cylinders, which were 3.18 m m in diameter and 8 m m in length, were provided by Zimmer Inc. (Warsaw, Indiana, U S A ) . The pore sizes o f these tantalum plugs were about 400 to 500 micrometers (shown in F i g .  31  4.1), and the porosity o f these porous tantalum was about 75%. The density o f solid tantalum was 16.65 g/cm . Weighted with a 1/100,000 g balance (Mettler A E 2 4 0 ) , the weights o f five 3  porous tantalum implants were listed in Table 4.1, and the average weight was 0.237 (SD 0.014) g.  Fig. 4.1 S E M images o f porous tantalum cylinder at various magnifications.  Table 4.1  The weight o f porous tantalum implant plugs  No. Weight (g)  1 0.225  2 0.237  4 0.228  3 0.261  5 0.235  average 0.237 (SD 0.014)  The surface area is a parameter, which affects current density, for porous tantalum plugs as substrates for electrolytic coating o f calcium phosphate. The estimation o f surface  32  area was based on the assumption that the porous tantalum plug was built with a single tantalum tube wrapped in cylindrical space o f the 3.18mm in diameter and 8mm in length. The dimensions o f this tube are 0.15 m m for outer diameter and 0.03 m m for inner diameter, as measured on B S E images. Based on the average weight and the density o f tantalum we could calculate the volume o f solid tantalum, which was then converted to the tube surface area. The surface area o f an implant plug thus estimated as 395 (SD 23.7) m m .  4.1.2 Electrolytic deposition (ELD) of calcium phosphate coatings 4.1.2.1 Surface cleaning of tantalum specimens Surface contamination such as grease and non-uniform oxidation o f metal substrates etc could affect coating kinetics, uniformity and morphology. Specimens needed to be cleaned properly before coating. The porous tantalum cylinders were ultrasonically bathed in acetone for 3 minutes and immersed in high concentration sulfuric acid (95 wt% H2SO4) (Fisher Science Inc) for 1 minute. The high concentration sulfuric acid can dehydrate and destroy the organic molecules. Both steps were designed to eliminate any organic contamination. The porous tantalum cylinders were then rinsed and cleaned i n ultrasonic bath of distilled water for 3 minutes, and then rinsed 5 times in distilled water to remove sulfuric acid (20ml or more water for more than 5 m i n each time).  The residue o f sulfuric acid from surface cleaning process should not effect bone cell reaction to implants. In animal or human bodies, there are about 0.3 m M / L sulphate in their serum [83], and these sulphate anions come from food, water and air. After rinsing, the  33  residue o f sulphate anions from implants should be much less than the sulphate anions in blood stream. "Excess sulphate i n the blood is rapidly eliminated by urinary excretion" [84].  4.1.2.2 ELD coating cell fixture A customized electrolytic cell fixture for electrolytic deposition was designed and manufactured in biomaterial laboratory (shown i n F i g 4.2). The fixed interval between 2 electrodes, which were cathode for implant plugs and anode for 25><25mm platinum plate respectively, is 9 mm. It was measured from the surface o f the platinum plate to the closer parallel tangent surface o f the porous tantalum cylinder. The anode clip was made o f stainless steel which was not immersed in the coating electrolyte, and the cathode rod which fully contacted with the porous tantalum cylinder was made o f tantalum. The porous tantalum cylinders were completely immersed in the electrolyte during E L D deposition.  Fig. 4.2 The image and schematic illustration o f E L D coating cell  34  4.1.2.3 ELD coating parameters In order to study calcium phosphate coating parameters, the crystal structure o f the coating, and the morphology and thickness o f calcium phosphate coatings, 100-micrometer thick tantalum foils o f 20x20mm and porous tantalum blocks o f 3 x 3 x 8 m m were used. Only half o f the square tantalum foil was immersed i n electrolytic solution, therefore the surface area o f this electrode was 4 0 0 m m which was close to the surface area o f porous tantalum 2  cylinders. The dimensions o f porous tantalum blocks were chosen to have a closer geometric shape to the porous tantalum cylinders.  The modification o f coating parameters was conducted based on the E L D coating technique developed by our group [61]. The parameters were selected to avoid the formation of hydrogen bubbles which harms the adhesion o f coatings, and the formation o f D C P D s which affects the uniformity o f coating structure and morphology. Other criteria, such as the cracking o f coatings, the thickness o f coatings and the uniformity o f thicknesses o f coatings at both inner and outer pores, which is particularly important for porous tantalum, are evaluated as well. The goal is to achieve coatings that are crack-free and have homogeneous micro-porous structure with a uniform thickness o f 3 to 5 micrometers o f calcium phosphate throughout the porous tantalum cylinder.  Some o f the coating parameters, such as the concentrations o f calcium and phosphate ions in the coating electrolyte, p H o f the solution and the applied voltage, and so on, have already been optimized [61] and were used without being changed in this study. Periodical stirring was introduced to increase the thickness uniformity across at central pores. The effect  35  of periodical stirring is examined by comparing coatings with and without this process. Coating time was chosen to achieve the maximum thickness o f coatings with minimum cracking. The modified processes o f E L D o f calcium phosphate are as follows.  The E L D o f calcium phosphate was carried out in an aqueous solution that contained calcium and phosphate ions which was prepared by adding proper amount o f calcium nitrate (Ca(N03)2-4H20), ammonium dihydrogen phosphate (NH4H2PO4) and sodium chloride (NaCl) (Fisher Scientific). The concentrations o f calcium and phosphate ions were 5.25 m M and 10.5 m M respectively, and the concentration o f Sodium chloride was 150 m M . The p H of the solution was then adjusted to 5.25±.07 by adding I N sodium hydroxide ( N a O H ) under the monitoring o f a calibrated p H meter (Thermo Orion, Beverly, M A . U S A ) . The applied D C voltage was 2.6 V , and the duration o f coating was 3 hrs. During the coating period, the coating solution was periodically stirred every 30 minute which should enhance the ion supply to the centre pores o f implant plugs.  4.1.3 Alendronate immobilization Bisphosphonates contain two phosphonate groups attached to a single carbon atom (C), forming a P - C - P structure. Alendronate (monosodium 4-amino-l-hydroxybutylidene1,1-diphosphonate trihydrate) was the bisphosphonate  used in the study. It was chosen  because o f its efficacy, availability, and clinical record.  The immobilization o f alendronate (Sigma Aldrich) was carried out by immersing calcium phosphate coated implants in 3.2 m l 10" M sodium alendronate in 0.01 M Phosphate 4  36  Buffer Saline (PBS) solutions at 37°C for 7 days. Proportion o f besphosphate ions from solution were adsorbed by the calcium ions on the Ca-P coating surfaces. Then, the soaked implants were rinsed with distilled water five times (20ml or more water for more than 20 m i n each time). The rinsing steps ensured that most o f the non-adsorbed alendronate ions were removed from implants.  4.1.4 Implant assemblies The gap model was achieved by installing polymethymethacrylate ( P M M A ) bone cement (Simplex ® P, Stryker Howmedica Osteonics, Limerick, Ireland) caps at both ends o f the implant plugs. The caps act as spacers between bone and the porous tantalum plugs. A s a result, the desired gap between the implant and the surrounding trabecular bone was maintained when the caps were press-fitted to the pre-drilled holes in the femoral bone. The final gap is a 0.6mm thick cylindrical wall with the inner diameter o f 3.18mm and an outer diameter o f 4.37mm. The volume o f the gap zone is about 35 m m . The details o f the 3  dimensions o f the final implant assemblies are illustrated in Fig. 4.3. Bone cement was used as the material for caps because it is clinically used in orthopaedic surgeries. A l s o , the flexible mixing ratio and adjustable curing time make it an ideal casting material.  37  Bone cement Spacing Cup  04.37mm - f — 03.18mm|-  Porous Tantalum Implant Cyfinder  Designed Gap  9mm  F i g . 4.3 Mechanical drawings o f porous tantalum implant assembly. The left is a 3 D drawing o f the implant assembly, and the right is a 2-D drawing with a l l the dimensions  Four customer-designed molds (Fig.4.4) were machined and used to cast bone cement caps. The ratio for bone cement was 0.7 grams o f powder per 0.5ml o f monomer liquid, which allows 5 m i n o f viscous state for handling. The dimensions o f the cap were 4.37mm in outer diameter, 3.18 m m i n inner diameter, 2 m m i n height and 1.5mm in depth. During assembly, the implant plugs were carefully handled to avoid any damage to the coating surfaces, and a drop o f bone cement (same ratio as above) was added into each cap as a glue to firmly attach the 2 caps to each end o f the implant cylinder (Fig. 4.3).  Fig. 4.4 The image o f bone cement casting mold  38  4.1.5 Sterilization of implants and surgical tools To minimize any damage to the rabbit's femoral bone, such as cracking during surgical drilling, a set of twist drill bits of serial diametric sizes was employed. They were 5/64" (1.98mm) brad tip, 1/8" (3.18mm) and 11/64" (4.37mm). These drill bits (Lee Valley Co. Canada) were ultrasonically cleaned in acetone and ethanol to remove organic contaminations. Then they were immersed in 70% isopropyl for pre-sterilization before being sealed in polyethylene bags (Food Saver). The implant assemblies were re-sealed in polymer packaging bags from Zimmer (see Fig. 4.5).  Fig. 4.5 Surgical tools and porous tantalum implant  Beta-ray radiation sterilization was carried out at Iotron Tech. Inc., Canada following the ISO13485:2003 international standard for medical devices. The applied radiation dosage was 25 to 27 KGys. Other tools such as the hand drill and surgical forceps were sterilized by autoclave at Jack Bell centre, Vancouver General Hospital.  39  4.2 High performance liquid chromatography (HPLC) tests for drug loading Five alendronate-immobilized calcium-phosphate-coated porous tantalum implants were tested for total amounts of drug loading. They were processed in the same way as the implants for animal tests. These 5 pilot samples were immersed in 1ml 1M hydrochloric acid, and the alendronate-containing calcium salt on the samples was dissolved in the acid solution. The tantalum samples were examined with the Nikon Eclipse optical microscope and the coating was confirmed to be completely dissolved. Seven alendronate solutions with known concentrations, which were 100, 50, 25, 12.5, 6.3, 3.1 and 1.6 ug /ml respectively, were also prepared as standard samples. These samples generated a linear calibration equation between alendronate concentration and a peak area of alendronate (IT = 0.9998).  During HPLC (high performance liquid chromatography) analysis, solutions reacted with 5 fold (V:V) of 2 mg/ml fluorescamine in acetonitrile to form a conjugate. The acetronitrile was extracted off with dichlormethane. The aqueous phase containing the conjugate was analyzed with a Waters chromatography station, with Nova Pac CI8 column and 470 scanning fluorescence detector. The conjugate was detected by excitation at 395 nm and emission at 480 nm.  4.3 Animal tests Alendronate-immobilized and calcium-phosphate-coated porous tantalum implants were tested in comparison with uncoated porous tantalum as control samples. To confirm the  40  effects o f alendronate on bone formation, calcium-phosphate-coated porous tantalum was adopted as another set o f control samples as well. The detailed plan is shown in Table 4.2.  Eighteen N e w Zealand White female rabbits o f 3.5-5.0 kg (28 to 34 weeks old) were employed for the animal tests. They were randomly divided into 3 groups for the 3 types o f implant assemblies which are non-treated porous tantalum (Ta) from Zimmer as control samples (6 rabbits), calcium-phosphate-coated porous tantalum (Ta-CaP) as second type of control samples (6rabbits), and alendronate-immobilized calcium-phosphate-coated porous tantalum ( T a - C a P - A L N ) (6 rabbits). The animal tests were conducted at the Jack B e l l centre (animal centre), Vancouver General Hospital. The study was approved by the A n i m a l Research Ethics Review Board at the University o f British Columbia (Protocol number: A 0 4 0275).  Table 4.2  A n i m a l test plan for bisphosphonate-coated porous T a bone implants  Ta  Ta-CaP  Ta-CaP-ALN  (Porous T a control)  (Porous T a with calcium Phosphate coating)  (Ta-CaP soaked in alendronate)  Rabbits  6  6  6  Number o f implants  12  12  12  Fluorescence labeled samples  12  12  12  Histological sections  36  36  36  S E M / fluorescene microscopy  36  36  36  Image analysis  36  36  36  41  4.3.1 Implantation procedures The implantation location o f the distal femur was selected because the trabecular bone inside the femoral distal end is similar to the bone at the pelvis. The rabbit was shaved and then anaesthetized by intramuscular injection o f a sleeping dose o f 3.15 to 3.5 mg ketamine/xylazine. The anaesthesia was maintained by inhalation o f 2 % to 2.5% isoflurane. Then, pain killer (0.3mg Buprenorphine, Temgesic) was administrated subcutaneously. The skin o f the rabbit, where the incision would take place, was cleaned with 0.5% chlorhexadine and then sprayed with 0.5% dexidin. These pre-surgery procedures were conducted by the animal care staff at the Jack B e l l Center.  The surgical procedure on the rabbits was performed by Dr. Winston K i m (a surgeon) in the animal operating room. A 3 cm incision was made on the distal lateral aspect o f the femur, and the vastus lateralis was split along its fibers to expose the underlying bone. One 4.37 m m hole was made bilaterally through sequential drilling (1.95 mm, 3.18 m m and 4.37 mm) under saline irrigation. The orientation o f the hole was perpendicular to the distal femoral condyle, and the depth was confirmed with a depth gauge. The porous implant with P M M A bone cement caps was gently inserted into the hole (see F i g 4.6). The wound was then irrigated and closed i n a uniform manner. T w o implant plugs with the same surface were implanted i n each rabbit, to avoid the confounding influence o f the presence or absence o f different implant coatings. The anaesthetic, operative procedure and animal care were performed in compliance with University and federal guidelines.  42  *orous Tantalum  Cement Cap  Condy!  / Trabecular Bone  G  a  p  Fig. 4.6 The images and schematic illustration of the implant location, a) digital image of implant position during surgery; b) X-ray image shows the implant position after surgery; c) the schematic illustration  4.3.2 Bone labeling Bone growth front was labeled with fluorescence dyes to track new bone formation with time. Two commonly used fluorochromes from Sigma-Aldrich, Alizarin complexone (absorption: 530-560 nm, emission: 624-645 nm) and calcein (absorption: 494 nm, emission: 517 nm), are dissolved in distilled water, and sodium bicarbonate was added to reach the concentration of 1.4% wt. The pH of the solutions was adjusted by adding hydrochloric acid if necessary. Within a Type II A bio-safety cabinet, these solutions were filtration sterilized with 0.2 micron filters (Millex GS, Carrigtwohill, Co. Cork, Ireland) and packaged in 10ml sealed syringes. The detail information about these two fluorochromes is listed in Table 4.3.  Table 4.3  Information o f fluorochomes and their solutions A l i z a r i n compleone A3882 530-560 nm 624-645 nm 30mg/ml 7.4  Sigma Catalog N o . Absorption Emission Solution Concentration Solution p H  The labeling chemicals were administrated subcutaneously.  Calcein C0875 494 n m 517 nm 1 Omg/ml 7.4  The labeling should  provide information about the direction o f bone growth, growth speed and bone growth mechanism. Calcein and Alizarin complexone were chosen because their emission wave lengths are within visual light range which means no special light sources such as ultra violet are needed. A l s o , these 2 chemicals are relatively mild to rabbits. A s shown i n Table 4.4, the dosages o f calcein and alizarin were 10 and 30milligrams per kilogram o f rabbit's body weight respectively.  Table 4.4 Calcein Calcein Alizarin Alizarin  Labeling administration dosages and schedule Dosage lOmg / K g lOmg/Kg 30mg/Kg 30mg/Kg  concentration 1 Omg/ml 1 Omg/ml 3 Omg/ml 3 Omg/ml  Solution volume 1 ml/Kg 1 ml/Kg 1 ml/Kg 1 ml/Kg  Day after surgery 7 day 14 day 2 1 day 2 8 day th  th  st  th  . 4.3.3 Animal care after surgery Antibiotic medicine, Notrile, was administrated after the surgeries. Rabbits were fed with carrots, hay and vitamins, and the weights and body temperature o f these rabbit patients were monitored regularly. Force feeding may be applied to those 'depressed' rabbits. Further administration o f antibiotics may be applied to those rabbits with higher body temperature.  44  4.3.4 Euthanasia, radiograph imaging and sample harvest After 29 days o f healing process, the rabbits were intravenously injected with Pentobarbital 2mg/Kg o f body weight to stop the heart beat i n less than 2 seconds. After the animal deceased, X-ray images were taken to reveal the implant positions. Afterward, the lateral legs were dissected, and then the femoral bones with the implants were removed for further histology analysis.  4.4 Sample analysis 4.4.1 Histology sample preparation The femur was cut parallel to the radius direction o f the condyle arc (see F i g . 4.7). The harvested samples were fixed, dehydrated, infiltrated and then embedded i n epoxy resin ( S P U R R , Canemco, St. Laurent, Q C , Canada) following a standard histological procedure (see Table 4.5) [85].  Table 4.5 Processes and schedules for fixation, dehydration and infiltration Processes Fixation Dehydration Dehydration Dehydration Dehydration Defat Infiltration Infiltration Infiltration Infiltration  Chemicals 10% buffered formalin 70% Ethanol/Water 90% Ethanol/Water 100% Ethanol 100% Ethanol 100% Acetone 50% SPURR/Acetone 80% SPURR/Acetone 100% S P U R R 100% S P U R R  45  Temperature 20°C 20°C 20°C 20°C 20°C 20°C 20°C 4°C 4°C 4°C  Duration 2days 3 days 3 days 3 days 3 days 3 days 3 days 3 days 3 days 3 days  Fig. 4.7 Schematic image for sectioning and observation directions  In order to examine bone formation at the site o f the implants and to reveal the 3-D structure, each embedded sample was longitudinally cut with a diamond saw (Buehler, Lake Bluff, I L U S A ) , and ground and polished (Buehler) into three sections at 200 um, 850 um and 1500 urn deep from the tangent surface o f the implant. Fig.4.8 gives the schematic illustration o f the three examining sections.  Section 1 Section 2 Section 3  Fig. 4.8 Schematic illustration for three examining sections  46  After samples were embedded in S P U R R (Canemco, St. Laurent, QC, Canada), the examining surfaces were ground to just touch the implant cylinders, which meant the grinding surfaces were the tangent surfaces o f the porous T a implant cylinders. The surfaces were considered the origin o f the coordinate. Then the 2 samples from same rabbit were reembedded together in Epothin (Buehler) resin i n a standard (1.25 inch i n diameter) mold. After curing, the bottoms o f the Epothin blocks were ground parallel to the sample surface with the grinding fixture at an accuracy o f ± 2 0 p m . The final surfaces were vibration polished with 0.05 micrometer silica slurry.  4.4.2 Histology and histomorphometry analysis 4.4.2.1 Fluorescence microscopy A fluorescence microscope (Eclipse E 600, N i k o n , Tokyo, Japan) with a F I T C & T R I T C dual exciter and dual emitter filter block, (exciter wavelengths: 475 - 495 nm and 540 - 575 nm, emitter wavelengths 500 - 535 nm and 580 - 620 nm), was used to analyze the fluorescence labels and new bone formation. The fluorescent imaging was performed before sputtering o f A u / P d alloy for S E M . Under the fluorescence microscope, bone formation front in the first 2 weeks was marked by two green calcein labels, while red alizarin Complexone lines marked bone formation in the 3rd and 4th week.  4.4.2.2 BSE imaging The back scattered electron ( B S E ) images were taken under a Hitachi-S3000N S E M (Hitachi Scientific Instruments, Tokyo, Japan). Polished samples were sputtered with A u / P d . B S E mode under high vacuum pressure was used. The beam current was set to 65mA,  47  voltage was 20 k V , and the aperture was removed. A l l these setups guarantee bone, epoxy and T a to have differentiable gray scales which meant the histogram would show 3 peaks.  4.4.2.3 Image analysis The image analysis was carried out on the low magnification B S E images using image analysis software (Clemex V i s i o n P E 3.5, Longueuil, Q C , Canada). The magnification calibration o f B S E images with image analysis software was 3.93 pm per pixel. The quadrilateral defined by the 4 corners o f caps was considered the sample area. Gray level discrimination was used to identify epoxy, new bone formation and implant. Figure 4.9 shows how Clemex processes the images. Analysis on the 3 sections o f B S E micrographs generated quantitative information on total gap volume, total volume available for bone ingrowths, and total volume available for new bone formation.  The volume o f bone gap filling, which was new bone formation in the gap region between the tantalum implant and host bone, and bone ingrowth into porous tantalum which was bone formation within the pores o f the tantalum plugs, were measured. The total bone formation (the sum o f the gap-filling and ingrowth) was then calculated. Total available tantalum length (see F i g . 4.10 (b): black lines) and the contact length (see F i g . 4.10 (a): blue lines) o f newly formed bone on the porous tantalum surface o f the three longitudinal sections of all implants were also measured. F i g 4.11 gave the schematic illustration o f the defined gap-filling and bone ingrowths in this study.  48  < 3  s  I  i 0  CQ  128  192  255  Fig. 4.9 Histogram for image analysis and the images before and after the dying process  49  Fig. 4.11 The schematic drawings illustrate the bone ingrowth and gap filling  4.4.2.4 Statistical data analysis Differences in the mean and standard deviation were investigated by a two-tailed ttest, using a statistical software program (SPSS Version 14.0, Chicago, I L , U S A ) . A n independent sample t-test was used to compare the volume o f gap filling, bone ingrowth, total bone formation, and the new bone formation contact length o f the different implants. We chose 95% for the confidence level and considered a p-value o f < 0.05 to be significant.  50  CHAPTER 5: RESULTS 5.1 The effects of coating parameters on calcium phosphate coating morphology M a n y coating parameters such as p H and the temperature o f the electrolytes, applied voltage, calcium and phosphate ion concentrations and coating duration affect the final coating crystal structure and coating morphology. Duan [61] has studied these parameters on flat titanium substrates and has optimized the coatings to a micro-porous structure good for drug delivery. To apply this E L D coating technique to porous tantalum substrates for the animal test, the coating process needs to be standardized to achieve repeatable coatings on all the alendronate-immobilized implants and the calcium-phosphate-coated controls. Some coating parameters need to be modified to fit the porous substrates o f tantalum.  The porous structure o f calcium phosphate coatings can be obtained on porous tantalum substrates by using the same concentrations o f calcium and phosphate ions, solution p H and applied voltage as those found on titanium plates. The remaining two major concerns on calcium phosphate coatings were cracking and spallation o f coatings and the uneven coating thicknesses at superficial pores and central pores.  Porous tantalum plugs were coated with calcium phosphate without sulfuric acid cleaning. The calcium phosphate did not uniformly cover the tantalum struts. O n some samples, uncoated spots surrounded by calcium phosphate coating could be seen (see Fig. 5.1 (a)). W i t h the sulfuric acid cleaning, these phenomena dramatically reduced.  51  S p a l l a t i o n a n d c r a c k i n g o f the p o r o u s c a l c i u m p h o s p h a t e  thickness  c o a t i n g are related to  o f t h e c o a t i n g i f o t h e r f a c t o r s , s u c h as c o a t i n g p o r o s i t y , r e m a i n c o n s t a n t .  d u r a t i o n w a s c h o s e n to c o n t r o l the c o a t i n g t h i c k n e s s  an optical microscope  w e r e a d o p t e d to e x a m i n e  in our experiments.  coating surfaces.  the  Coating  Observations using  T h e d u r a t i o n o f the E L D  coating process was determined; hence few cracks or crack-free coatings were achieved. F i g .  5.1  (b) s h o w s s p a l l a t i o n h a p p e n i n g o n a p o r o u s t a n t a l u m s a m p l e . T h e c o a t i n g d u r a t i o n f o r  p o r o u s t a n t a l u m i m p l a n t s w a s shorter t h a n that for this  F i g . 5.1  S E M images o f calcium phosphate  coatings,  sample.  a) w i t h o u t  sulfuric acid cleaning,  p o r o u s t a n t a l u m strut w a s n o t c o a t e d ; b ) s p a l l a t i o n o f c a l c i u m p h o s p h a t e c o a t i n g s ; c ) morphologies  at  central  pores  without  periodical  stirring;  superficial l a y e r w h e n p e r i o d i c a l stirring w a s not a p p l i e d .  52  (d)  coating  the  coatings  morphology  at  Several porous tantalum plugs were coated with calcium phosphate without periodical stirring. F i g . 5.1 (c) and (d) show coatings at the central pores and at the superficial pores on one sample respectively. A t the central pores (Fig. 5.1 (c)) the coating was very thin (less than 500 nm), and the sharp edges o f the metal grains could be seen. A l s o , the pore size of calcium phosphate coatings was very small. A t the surface (Fig. 5.1 (d)) o f the porous tantalum cylinder, larger pores o f calcium phosphate coatings were formed, and the metal grain edges were blunt or buried under calcium phosphate coating. The estimated thickness was more than 3 micrometers.  The causes o f these uneven coating thicknesses in Figure 5.1 might be the insufficient supplies o f calcium and phosphate ions to the central pores because all ion supplies defusing from the surrounding solution were consumed by the superficial layer. Periodical stirring was introduced to the coating process to manually deliver the calcium and phosphate ions into the central pores o f porous tantalum as the ion supplies for coating. F i g . 5.2 shows the improved coating at a central pore with the periodical stirring.  Fig. 5.2 S E M images on calcium phosphate coatings at a central pore when periodical stirring was applied.  53  5.2 Calcium phosphate coating and alendronate loading 5.2.1 ELD coating of calcium phosphate The digital image and S E M images under different magnifications (Figure 5.3) show the morphology o f calcium phosphate coatings on porous tantalum implants. The digital image (Fig. 5.3(a)) shows the complete implant assembly. The low magnification image (Fig. 5.3(b)) shows that the coverage o f the calcium phosphate coating over the complex 3D tantalum structure is almost 100%. Higher magnification images (Fig. 5.3(c), (d)) illustrate the coating morphology which is a thin layer o f calcium phosphate with a micro-scale porous structure (about 1 um in pore size).  A tantalum foil sample was employed to conduct a coating thickness measurement. The coating parameters for this foil substrate were the same as those for porous tantalum cylinders. After coating, the foil was bent 180° to crack the calcium phosphate coatings. The coating thickness was 5.8 micrometers as measured on an S E M image (Fig. 5.4).  54  Fig. 5.3 S E M images at varying magnifications o f Ca-P coating morphology, a) digital image o f implant assembly; b) low magnification S E M image shows the coverage o f C a P coating; c) and d) the porous morphology o f C a P coating.  5.2.2 The alendronate loading amount and estimation of dosage The H P L C experimental results o f alendronate loading amount on the 5 pilot porous tantalum plugs were calculated, and the drug dosages were estimated accordingly. The total drug amount loaded by immobilization was at microgram level, and the average drug amount on these 5 porous tantalum plugs was 1.37 ( S D 1.0) micrograms. The molecular weight o f alendronate was 325.12 grams per mol. The volume o f gap was calculated as - 3 5 m m , and the volume o f pores was calculated as - 4 8 m m , based on the dimensions and measured 3  porosity o f porous tantalum cylinders. The total volume o f sample zone was - 8 3 m m .  Assuming that all drugs dissolved in one day, which was the extreme case that was unlikely to happen, the highest possible drug dosage was -4.7 x l 0 " M / L . This dosage was 5  still in the effective zone o f alendronate concentrations which were from 250x10"  9  to 10"  4  M / L per day according to Bergstrom[86]. In our animal tests, there should not be any concern o f overdose o f alendronate. If all the drugs and calcium phosphate slowly dissolved in 20 days, the average daily dosage should be ~2><10" M / L which was within the effective 6  concentration zone.  5.3 Crystal structure of E L D calcium phosphate coating X R D results showed the E L D coatings to be a mixture o f H A and O C P (Fig. 5.5), as evidenced by the diffraction (shoulder) o f O C P at 29 = 4.72° and diffraction o f H A at - 5 0 (49.46 (2 1 3) plane). The standard H A or O C P has higher intensity at 33 degrees than that at 26 degrees. The X R D spectra o f E L D coatings (and alendronate-immobilized coatings) have higher intensity at 26 degrees ((002) planes), indicating a preferred orientation o f the calcium  56  phosphate crystals in E L D coatings. Comparing the two X R D spectra, it can be concluded that alendronate immobilization did not change the crystal structure o f the calcium phosphate coating.  HA/OCP (002)  cz CD  After  Before  10  20  30  40  50  2e (degree) Fig. 5.5 X R D results for CaP coatings before and after alendronate immobilization  F T I R spectra o f the coatings before and after alendronate immobilization are shown in Fig.5.6. The PO4 " peaks at 561 and 600 cm" are close to those o f octacalcium phosphate 1  3  (OCP), which is one type o f calcium phosphate crystal structure with PO4 " peaks at 561 and 3  599 cm" [87 ]. A s a comparison, the PO4 peaks o f hydroxyapatite ( H A ) around 600 cm" are 566.7 and 603 cm" [88], and the interval between these two peaks is smaller than the current 1  result. After alendronate immobilization, F T I R spectrum shows slight changes at 1080 and - 8 7 5 cm" . F T I R spectrum o f alendronate calcium salt shows the P - 0 stretching modes o f 1  phosphonate at 1090, 1000 and 964 cm" [ 8 9 ] . The changes i n F T I R spectrum after 1  57  immobilization might be because o f the chelation o f alendronate and calcium ions on the calcium phosphate coating surfaces.  The energy dispersive X-ray spectrometry ( E D S ) tests on E L D calcium phosphate coatings were carried out on tantalum foil substrates to compare the coatings o f calcium phosphate before and after alendronate immobilization. The atomic ratio o f calcium and phosphorus can be used as supporting evidence o f the crystal structure because different calcium phosphate crystals have different Ca/P ratio. O n each type o f coating, five different areas located at the centre and the 4 corners were tested, and the averages o f CaP atomic ratios were calculated to be 1.54 and 1.53 before and after alendronate immobilization respectively. The adsorption o f alendronate onto the coating thus did not significantly change the Ca/P ratio o f the calcium phosphate coating.  1400  1200  1000  800  600  400  Wavenumber (cm ) 1  Fig. 5.6 The F T I R results for CaP coatings before and after alendronate immobilization  58  In conclusion, the crystal structure o f the calcium phosphate coatings was not changed during alendronate immobilization based on the results o f X R D , F T I R and E D S . The coatings before and after alendronate immobilization are predominantly O C P . However, the existence o f H A can not be excluded. The crystals in the E L D coating are oriented.  5.4 Histology The microscopic analysis o f peri-implant bone growth was carried out with a fluorescence  optical microscope and a scanning electron microscope. The bone growth  patterns, including bone formation, growth directions and the bone/implant interlocking etc, were studied. The comparisons o f these histological results between control groups and the drug-loaded group were conducted.  Figure 5.7 shows the B S E images (left column) and fluorescence optical images (right column) o f bone growth on the 3 types o f implants: Ta control (Fig.5.7 (a), (b)), TaCaP control (Fig.5.7 (c), (d)) and T a - C a P - A L N implant (Fig.5.7 (e), (f)). In the B S E images, it could be obviously seen that peri-implant bone growth and ingrowth into the implant were the highest on the T a - C a P - A L N sample (Fig.5.7 (e)). Between the 2 types o f control samples, the Ta-CaP control had more bone growth in defined sample area than the T a control did.  59  c  4  d  2 mm f  /  * • ft  2 mm  2  mm  F i g . 5.7 B S E and fluorescence images o f bone formations on 3 types o f implants, a) and b) non-coated control; c) and d) calcium phosphate coated control; e) and f) alendronateimmobilized calcium-phosphate-coated implant.  The fluorescence optical images i n Figure 5.7 show the whole bone sections where the sectioning planes were made along the axis o f implant cylinders. The shining green colour was the emission light o f calcein under the excitation light o f 495 nm wavelength.  60  Again, we could observe significant increases o f bone growth on T a - C a P - A L N implant comparing with those on the 2 types o f controls. Also;" the pore size o f the newly grownperiimplant bone was smaller in comparison with two other control samples. Furthermore, the t green colour indicated that these new bones were formed in the first 2 weeks post-surgery. The depth o f ingrowth into porous tantalum implants on T a - C a P - A L N sample was higher than those on the two types o f controls in which ingrown bone only reached the superficial layer o f pores.  5.5 Histomorphometry The ingrowth was the new bone that grew into porous tantalum plugs; the gap  filling  was the new bone growth in the designed gap area, and the total bone growth was the sum o f the ingrowth and the gap filling. The information from 3 sections was integrated into 3-D volume percentages. For the bone/implant contacts, the data were presented separately.  5.5.1 Bone formation 5.5.1.1 Gap filling The mean extent o f gap fillings was 13.0 % (SD 5.0 %) for T a controls, 15.8 % (SD 3.3 %) for T a - C a P - A L N controls, and 29.1 % (SD 10.0 %) for T a - C a P - A L N implants (shown i n F i g 5.8). The relative increases in volume o f gap fillings on T a - C a P - A L N implants were 124% (2.24 fold) and 84% (1.84 fold) compared with T a controls and Ta-CaP controls respectively.  61  The t-tests, comparing T a - C a P - A L N  implants with Ta controls, indicated the  difference o f gap fillings between these two groups was significant (p-value was less than 0.0001) (see Table 5.2). Between T a - C a P - A L N implants and Ta-CaP controls, the difference o f gap fillings was significant as well (p-value was less than 0.0005). O n the other hand, the calcium phosphate coating slightly increased the bone formation volume o f gap filling, but the p-value o f t-test was 0.12 which meant that the difference o f gap fillings between these two groups were not significant.  5.5.1.2 Bone ingrowth The mean volume o f bone ingrowth was 6.0 % ( S D 2.7 %) for T a controls; 8.3 % (SD 2.9 %) for Ta-CaP controls, and 19.9 % (SD 5.9 %) for the T a - C a P - A L N (see Fig.5.8). The increases i n bone ingrowths on T a - C a P - A L N samples were 232% (3.32 fold) and 140% (2.40 fold) compared with T a controls and Ta-CaP samples respectively.  The p-value o f the t-test in comparison o f T a - C a P - A L N implants with T a controls resulted in less than 0.0001 which meant the difference o f ingrowths between these two groups was significant (see Table 5.1). The difference o f ingrowths was also significant between T a - C a P - A L N implants and Ta-CaP controls (p-value was less than 0.0001). In comparison to T a controls, bone ingrowth was slightly increased on Ta-CaP samples, but the p-value o f the t-test was 0.07 which meant the increase o f bone ingrowth on Ta-CaP samples was not significant.  62  5.5.1.3  Total bone formation The volume o f total bone formation, which was the sum o f gap filling and bone  ingrowth, was 9.7 % ( S D 3.5 %) for T a controls, 12.4 % (SD 3.2 %) for Ta-CaP controls and 26.2 % (8.3 %) for T a - C a P - A L N implants. The increase o f total bone formation on Ta-CaPA L N implants was 170 % (2.7 fold) compared with that on Ta controls. The increase was significant because the p-value o f t-test was less than 0.0001. The total bone formation on T a - C a P - A L N implants showed a 111 % (2.11 fold) increase in comparison to Ta-CaP samples, and it was also statistically significant (p < 0.0001).  60  Ta 50  40  |  o >  30  W&  -  -  ** *  I  Ta-CaP I Ta-CaP-ALN  ** *  ** *  —  20  10  0  G a p Filling  Bone Ingrowth  Total Bone Formation  Fig. 5.8 The mean and standard deviation o f volume o f gap filling, bone ingrowth and total bone formation. T a - C a P - A L N implants demonstrated significantly increased gap filling, bone ingrowth and total bone formation compared with Ta-CaP controls (p < 0.0005, p < 0.0001 and p < 0.0001 respectively *), and significantly increased gap filling, bone ingrowth and total bone formation compared with T a control (p < 0.0001, p < 0.0001 and p< 0.0001 respectively **). N o significant differences in gap filling, bone ingrowth and total bone formation were found between Ta-CaP and Ta. (p = 0.12, p = 0.07, and p = 0.07).  63  Table 5.1  The p-values o f t-tests on bone formation, gap filling and ingrowth  Compared groups Ta & Ta-CaP Ta & T a - C a P - A L N Ta-CaP & T a - C a P - A L N  Bone formation, v % 0.07 < 0.0001 < 0.0001  Gap filling v % 0.12 < 0.0001 O.0005  Ingrowth v % 0.07 < 0.0001 < 0.0001  5.5.2 Bone/implant contact length The direct contact between newly grown bone and the implant is a very important factor that affects the mechanical fixation o f implants. This is especially critical when there is a gap existing between the implant and its surrounding host bone, the contacts between newly formed bone and the implant directly dictate the implant fixation.  In the 2 D sections, the contacts between bone and implant only encounter new growth bone because the gap model design avoids any press-fit contacts o f host bone and implants. A l s o , the bone/implant contact interfaces in 3D structure becomes the border lines between bone phase and implant metal phase in the 2 D images. The length o f these contact border lines directly represents how well the new bone contacts the implants. Therefore, the bone/implant metal contact length can be used as a criterion to judge the bone/implant contact and the implant fixation.  The mean percentages o f bone-implant contact length over total available tantalum length o f the three sections were measured (Fig 5.9). In section 1 (200 micrometer deep from tangent surface o f implant), the percentage o f contact length was 8.6% (SD 3.9%) for the Ta controls; 13.2% (SD 4.9%) for Ta-CaP controls and 52.3% (SD 18.3%) for T a - C a P - A L N implants. In section 2 (850 micrometer deep), the percentage contact length was 3.8% (SD  64  2.1%) for T a controls;..8.1% (SD 2.8%) for Ta-CaP controls and 39.4% ( S D 14.3%) for TaC a P - A L N implants. In section 3 (the central surface o f implant), the percentage contact length was 4.6% ( S D 1.4%) for T a controls; 9.0% (SD 3.0%) for Ta-CaP controls and 40.0% (SD 17.6%) for T a - C a P - A L N implants. The percentage o f contact length o f T a - C a P - A L N implants increased by an average o f 700% (8-fold) and 342% (4.42 fold) compared with those o f T a and Ta-CaP controls respectively. The average increase o f contact for Ta-CaP controls was 87% (1.87-fold) compared with that o f T a controls.  The t-test results show that the enhancements o f bone/implant contacts at the site o f T a - C a P - A L N samples were significant compared with Ta and Ta-CaP controls. The pvalues were less than 0.0001 in both cases. A l s o , the increase o f contact at the site o f Ta-CaP controls was significant compared with T a controls because the p-value oft-test was 0.02.  Table 5.2  The p-values o f t-tests on bone/implant contacts o f 3 sections  Compared groups T a & Ta-CaP Ta & Ta-CaP-ALN Ta-CaP & T a - C a P - A L N  Section 1 0.02 O.0001 O.0001  65  Section 2 0.0004 O.0001 O.0001  Section 3 0.0002 O.0001 O.0001  C? 100 Ta CO  •£ c  80  OD  i  O  Ta-CaP Ta-CaP-ALN  ** *  <D  ** *  CL  £ 60 CT) c  d) _l o CO c o O  *  40  20  I  0  c o  Section 1  Section 2  Section 3  Fig. 5.9 Bone formation on the surface o f porous tantalum expressed as a percentage o f the available porous tantalum surface (contact length) o f three sections performed through each implant. T a - C a P - A L N demonstrated significantly increased contact length compared with Ta-CaP controls, p < 0.0001 *, and significantly increased contact length compared with Ta controls, p < 0.0001 **. Ta-CaP controls demonstrated significantly increased contact length compared with T a controls, p < 0.02 +.  5.6 Bone formation mechanism Fluorescence labeling provides the information on when and where the  bone  formation happens and the directions o f bone growth. Base on the fluorescent images, some facts regarding the bone formation mechanism were observed and the comparison among 2 control groups and T a - C a P - A L N implant group was conducted. F i g . 5.10 shows how the fluorescence labeling works. F i g . 5.10 (a), which is the transverse section o f a tooth, shows the fluorescent rings. The growth direction can be identified by these rings with reference to the administration sequence o f fluorochromes (see Table 4.4 on page 45). F i g . 5.10 (b) shows the bone initiation on a Ta-CaP control surface.  66  A l l samples (108 sections from 36 implants) revealed the new bone formation initiated at the surfaces o f the host bone and bridged over the gap towards metal implants during the first 2 weeks after surgeries. This is called the distance osteogenesis [26], and the distance osteogenesis w i l l occur regardless o f the types o f implants because the host bone is the best places for new bone formation.  F i g . 5.10 Fluorescence images showing the labels on tooth and ingrowth bone, and the ordinal numbers gave postoperative time sequence, a) 4 color rings indicated the growth direction on tooth. The 2 green rings were calcein and 2 red rings were alizarin; b) labels indicated the bone initiation on implant surface.  The major difference between T a - C a P - A L N implants and 2 control groups was the interaction o f newly formed bone with the implant surfaces. The pattern and site o f bone initiation were strongly modulated with alendronate-immobilized calcium phosphate coating. In the T a controls, bone formation occurred inside the tantalum pores with little attachment to the surface o f porous tantalum. F i g . 5.11(a) shows the bone initiation (the green lines)  67  occurred around the tantalum strut (not touching) and expanded (the red lines) afterward. In contrast, early bone formation on T a - C a P - A L N plugs occurred predominantly on the surface o f the tantalum struts. In F i g 5.11(b), the new bone initiation (green lines) spread on the implant surface and grew outward. Ta-CaP controls had a similar pattern o f bone formation to that o f the T a controls, and bone initiation on implant surfaces was occasionally observed on the Ta-CaP controls. F i g 5.11 (c) shows similar bone initiation on a Ta-CaP control to that on a T a control. Fig. 5.11 (d) demonstrates the bone initiation on the implant surface o f a TaCaP control. But the initiation o f bone on Ta-CaP controls was slightly different than that on the T a - C a P - A L N implants. It only covered a portion o f implant surface and expanded into pore center.  The bone ingrowth patterns were different on T a - C a P - A L N implants from those on the two types o f controls because o f the change in bone formation initiation. O n T a controls (see Fig. 5.12), early bone formation occurred i n the center o f the porous area and expanded outwards onto the porous tantalum struts, as was tracked by the fluorescence labels (green lines were for first and second postoperative weeks, and red lines were for third and fourth weeks). The bone ingrowth went through the channel o f tantalum pores and barely touched the struts. O n Ta-CaP controls (see F i g . 5.13), the similar ingrowth pattern to that o f Ta controls could be seen, and also a branch o f bone ingrowth anchored on implant surface (Fig 5.13 (b) left bottom corner). O n T a - C a P - A L N implants (see F i g . 5.14), the bone ingrowth spread on the implant surface and followed by bone expansion towards the center o f the porous channels.  68  Fig. 5.11 Fluorescence images showing bone initiation, a) non-coated porous tantalum control; b) alendronate-immobilized calcium-phosphate-coated implant; c) and d)calciumphosphate-coated controls.  69  Fig. 5.12 Fluorescence images showing bone ingrowth pattern on T a control, a) low magnification image; b) higher magnification image showing the labeling lines.  Fig. 5.13 Fluorescence images showing bone ingrowth pattern on Ta-CaP control.  F i g . 5.14 Fluorescence images showing bone ingrowth pattern on T a - C a P - A L N implant.  70  The changes i n bone formation mechanism also affected the patterns o f gap filling bone growth. O n the T a controls (see F i g . 5.15 (a)), new bone formation crossed over the gap area but did not anchor on the implant surface; on the Ta-CaP controls (see F i g . 5.15 (b)), there were two small portions o f new bone anchored on the implant surface; on T a - C a P - A L N implants (see F i g . 5.15 (c) and (d)), new bone fully contacted the implant surface.  a  b  Cap — • !  tiap  ~~1  la  ^^^^ Ta  '  1V  *  300 u m  1  300 u m  Fig. 5.15 Fluorescence images showing the bone formation at gap areas. The white lines with double arrows indicated the bone bridging gaps a) T a control; b) Ta-CaP control; c) and d) T a - C a P - A L N implants.  It is very important for bone/implant fixation that the new bone crossing gaps anchored on implant surfaces. Without the full bone/implant contact, there would be a possibility o f micro-motion between implant and host bone. Without the bone formation  71  initiation on implant surface, the bone/implant integration would be postponed because the bone initiation on implant surfaces ensured a certain level o f bone/implant integration in first 2 weeks postoperatively on T a - C a P - A L N implants compared with bone still barely touching the implant surfaces 4 weeks after the surgeries on the T a controls.  72  CHAPTER 6: DISCUSSION 6.1 The effects of calcium phosphate coating Hydroxyapatite is a bioactive ceramic. M a n y reports on hydroxyapatite enhancing bone formation were published [90, 91,92].  Barrere et al [93] claimed that biomimatic  coating o f O C P increased the osteoinductivity (defined on page 12) o f porous tantalum when implant locations were inside the back muscles goats, and he could not fully claim the increase o f osteoinductivity by biomimatic coated O C P when the porous tantalums were implanted i n distal femur. Davies J E [23] believed the B M P might adsorb on porousstructure hydroxyapatite and therefore enhance osteoinduction. However, with a.gap existing between implants and host bone, our animal tests did not reveal significant increase o f periprosthetic bone formation. Although there were 21.5% and 38.3% increases o f gap filling and ingrowth respectively, the p-values o f t-tests were 0.12 and 0.07 respectively. Therefore, the osteoinductivity o f porous tantalum implants was not significantly increased by calcium phosphate coating process when challenging bone/implant gaps and defects.  Despite the weak effects on the osteoinductivity o f implants, the calcium phosphate coating significantly increased the new bone-formation/implant contacts. Our testing results showed an average o f 87% (1.87-fold) increase o f contact lengths on Ta-CaP implants compared with those on Ta controls. W i t h the help o f fluorescence images, the initiation of new bone formation on the strut surfaces, which was also called contact osteogenesis, was occasionally observed on Ta-CaP implants compared with little bone initiation on implant  73  surfaces o f T a control. These results provided the direct evidences and explanations to why bone/implant contacts were increased by calcium phosphate coating process.  The initiation o f new bone formation on calcium phosphate coating surfaces indicated better attachment o f osteoblasts and osteogeneritor cells. In another words, the calcium phosphate coating process increased the osteoconductivity (defined on page 12) o f porous tantalum implants. Theoretically, the increase o f bone /implant contact may contribute to the eventual implant fixation.  Summer et al [46] reported that the hydroxyapatite-tricalcium phosphate coating did not contribute to fixation, bone ingrowth, and bone formation based on their mechanical testing and histomorphometry results o f a gap model animal test on B M P treated implants. They analyzed the fixation strength, interface stiffness and energy to failure and found no difference with or without hydroxyapatite-tricalcium phosphate coatings. They also analyzed the bone volume, trabecular numbers and trabecular plate thickness etc on both calcium phosphate coated samples and the controls. They found no difference either. Their results supported our findings.  In conclusion, in the presence o f a gap between bone and the implant, our E L D calcium phosphate coating increases the bioactivity o f porous tantalum implants in terms o f osteoconductivity. The testing results show evidence o f better osteoconductivity and contact osteogenesis on calcium phosphate coating surfaces compared with non-coated tantalum surfaces. However, the calcium phosphate coating treatment did not significantly change the  74  osteoinductivity o f porous tantalum, and therefore the peri-prosthetic new bone formation was not increased. A s a result, the bone/implant fixation might not be obviously improved by calcium phosphate coating process.  6.2 The effects of alendronate-immobilized calcium phosphate coatings Our gap model study demonstrated a significant enhancement o f bone ingrowth (19.9 % versus 6.0%), bone gap filling (29.1% versus 13.0%) and total bone formation (26.2 % versus 9.7%) associated with T a - C a P - A L N samples compared with T a controls. A t the same time, an alendronate surface treatment significantly enhanced bone ingrowth (140 % ; 2.40fold), gap filling (84%; 1.84-fold) and total bone formation (111%; 2.11-fold) in comparison with Ta-CaP controls.  Possible explanations o f the enhancement o f bone formation towards implants could be: the effects o f alendronate on bone cells increased the activities o f osteoblasts work and decreased the activities o f osteoclasts; or the alendronate-immobilized calcium phosphate coating increased osteoinductivity o f porous tantalum which meant more stem cells differentiated to osteoblasts. Davies J E [23] believed the osteoblasts would be trapped in bone and became osteocytes after the mineralization o f the bone matrix. For further bone growth, new osteoblasts needed to be recruited. In order to clearly understand the mechanism behind the enhancement o f new bone formation, further studies need to be conducted.  The  histomorphometry analysis also showed that the alendronate-immobilized  calcium phosphate coating significantly increased bone/implant contacts compared with Ta-  75  CaP controls (342%; 4.42-fold) and T a controls (700%; 8-fold). The fluorescence imaging found that the initiation o f bone formation happened on the implant surface throughout the T a - C a P - A L N implants. The osteoconductivity o f these implants was therefore dramatically increased compared with both Ta-CaP and T a controls. The significant increase o f contact osteogenesis on alendronate-immobilized calcium phosphate coating indicated that it was the best bioactive implant surface out o f the three types o f implants.  The X R D , F T I R and S E M imaging analysis o f coatings revealed no changes on morphology and crystal structure o f the E L D calcium phosphate coating after alendronate immobilization. This eliminated any other potential factors, such as a change in the crystal structure o f calcium phosphate from O C P to H A , that happened and affected the bioactivity of the coating surfaces. Therefore, the dramatic enhancements  o f bone formation and  osteoconductivity o f porous tantalum implants were essentially due to the presence o f alendronate.  The results o f our study are also consistent with recent in vivo tests reported in the literature. The effect o f zoledronate on bone ingrowth into porous tantalum implants was studied i n a canine model by a single post-operative intravenous injection [67]. Subsequently, zoledronate was bound to hydroxyapatite-coated porous tantalum, and an increase o f 134% in bone ingrowth was reported by the. same group [68]. In our gap model with a less potent bisphosphonate  (alendronate), however, we found a greater relative increase in bone  ingrowth compared with the zoledronate study (232% versus 134%) [68].  76  V  The differences in our findings may be related to several factors including differences in animal models, the type o f bisphosphonate used, length o f implantation, location o f implantation and the design o f the implants. Another key difference between our study and previous studies is the unique micro-porous calcium phosphate coating made by E L D technique. Such a coating may not significantly affect bone formation itself, but it acts as a very effective carrier for bisphosphonate delivery.  Because o f the effects o f alendronate-immobilized calcium phosphate coating on osteoconductivity o f porous tantalum implants, bone reaction to implants was completely changed, and the initiation o f bone formation happened on the surface o f the alendronateimmobilized calcium phosphate coatings. In addition, alendronate-immobilized calcium phosphate coating increased new bone formation towards implants. A s a result, more bone formed bridging the gap between host bone and implant and anchored onto the implant surface, and eventually better bone/implant fixation could be achieved.  6.3 Clinical relevance of the gap model animal tests and alendronate delivery implants It has been known that biological fixation o f porous implants occurs by microinterlocking with ingrown tissue. Autopsy studies have demonstrated that bone ingrowth occurs in only 30-40% o f the surface area o f the implant in primary total hip arthroplasty [94]. In revision hip arthroplasties, bone defects and gaps may significantly compromise the quantity and quality o f host bone available for ingrowth. In many cases less than 50% host  77  bone is available for bone ingrowth o f the newly implanted shell. Therefore, any solutions to achieve a reliable and durable bone ingrowth to porous materials would potentially lead to novel implants for challenging revision cases.  Gaps and defects that occur between the surface o f the implant and the host bone in an acetabular reconstruction are usually filled with bone graft or bone graft substitutes in current clinical practice. However, bone ingrowth into porous implants is poor in the presence o f acetabular defects, regardless o f the defect filling with bone graft. In an earlier study, acetabular defects were created i n a canine model which were left unfilled or were filled with either autograft  bone or a 50:50 mixture o f autogenous bone graft and  hydroxyapatite/tricalcium phosphate ( H A / T C P ) ceramic [95]. A t 6 weeks post operation, grafting improved defect healing but had inferior bone ingrowth compared with empty defects.  In our study, a rabbit implant model incorporating a gap between porous tantalum and bone was developed to assess the effect o f alendronate coated porous tantalum compared against the controls. The model sought to replicate the clinical scenario in total hip revision arthroplasty. Implantation o f the plugs in the metaphyseal region also closely replicates the clinical scenario in that the implants were placed in cancellous bone. The positive results o f our study indicate that the surface treatment technique could provide an effective solution to the clinical challenges discussed above. Another application is the spinal fixation where fast and more bone ingrowth to the implants are critical to the function.  78  6.4 Key progresses in ELD surface drug delivery and animal studies A unique technique i n this study is the electrolytic deposition and the resulting microporous calcium phosphate coating. The calcium phosphate coating itself does not seem to have significant impact on bone growth. This relative uniform calcium phosphate coating, however, provides an ideal surface carrier for bisphosphonate local delivery. The electrolytic deposition used in this study has obvious advantages over plasma spray and biomimetic coating for porous T a implants. In plasma spray, only a half thickness o f the porous cells on T a implant surface can be coated with calcium phosphate because o f the line-of-sight limitation. A s a result, bone ingrowth into deeper porous area can not be improved. Biomimetic coating technique is a relatively slow process. It requires careful surface pretreatment and strict control over the solution conditions. The calcium phosphate coatings tend to form spherulites. Electrolytic deposition with periodical stirring could make a relative uniform and porous coating both inside and on the surface o f the porous T a implants.  Another unique feature o f our animal tests is the introduction o f fluorochrome injections to the bone formation mechanism studies. The fluorochromes mark the front o f bone formation, and many people have used a series o f fluorochromes to reveal bone growth direction and growth speeds etc. W e employed this fluorescence technique to examine the mechanism o f bone formation on different types o f surfaces o f porous tantalum implants and visualize the initiation o f bone formation on the surface o f alendronate-immobilized calcium phosphate coating.  79  Alendronate-immobilized calcium phosphate coating provides an advanced drug delivery method that can directly send alendronate to symptom location. It dramatically reduces the drug dose without compromising any healing effects. A l s o , alendronate, as one o f bisphosphonates, has many advantages over many other materials such as hydroxyapatite and B M P . Other methods o f augmenting the rate and extent o f bone ingrowth into porous implants have been described in the literature, including the use o f B M P [46, 96], Bone-like Carbonated Apatitic ( B C A ) coating [93], bone matrix gel [97] and non invasive low intensity ultrasound [98]. The advantage o f bisphosphonates over B M P is their relatively low cost. Other advantages o f bisphosphonates are their established clinical record, their effects on reducing bone resorption thereby decreasing stress shielding [99] and wear particle-induced osteolysis [41].  Improvements to the design and biomaterial properties o f any drug-implant delivery system can only be made by understanding the principles underlying the mechanism o f action o f the drug- delivery system. The mechanism o f enhanced bone ingrowth and osteoconductivity o f T a - C a P - A L N provides further insight into the underlying action o f this drug  delivery system.  Enhancement  o f bone  ingrowth, gap  filling  and  increased  osteoconductivity o f T a - C a P - A L N could have very significant advantages in revision joint arthroplasties associated with host bone defects, providing early and enhanced biological fixation i n complex cases.  80  CHAPTER 7: CONCLUSIONS A gap model animal test was designed and successfully conducted to study the effects of alendronate-immobilized calcium phosphate coatings on bone formation towards porous tantalum implant. The gap model imitated the clinical scenario. This and other unique features, such as the porous E L D coatings o f calcium phosphate, alendronate immobilization and fluorescence labeling etc, distinguished this animal test from any others and represented significant  improvements in implant/bone interaction study. Following are the main  conclusions drawn from this animal test: 1.  Alendronate-immobilized calcium phosphate coating enhanced the bone formaton towards porous tantalum implants. The treatment could significantly accelerate early stage (4 weeks) bone growth (by 170%) as compared with porous T a by itself. The increase was 111% when compared with Ta-CaP controls. Although Ta-CaP implant increased total bone formation by 28%, the improvement was not statistically significant.  2.  Both calcium phosphate coatings before and after alendronate significantly  increased  the  Alendronate immobilized  osteoconductivity  calcium phosphate  o f porous  immobilization  tantalum  implants.  coating achieved much  greater  enhancement. Alendronate immobilization treatment demonstrated dramatic and significant increase in bone/Tantalum contact. A s compared with porous tantalum itself, T a - C a P - A L N implant increased bone coverage on tantalum by 770%. The increase was 344% when compared with Ta-CaP controls. Calcium phosphate  81  coating increased the bone/implant contact length by 87% compared with T a controls.  3.  The observation o f fluorescence labeling images showed that the treatment o f alendronate-immobilized ELD-coating o f calcium phosphate changed the new bone formation mechanism. N e w bone formation initiated on the T a - C a P - A L N surfaces, and this provided the explanation why new bone/implant contacts increased on TaC a P - A L N implants.  4.  Early bone/implant integration may be achieved on T a - C a P - A L N implants. The fluorescence images showed that new bone crossing gaps anchored on the implant surface in the first two weeks postoperatively.  5.  Alendronate-immobilized calcium phosphate coating on porous tantalum  has  advantages in overcoming the gaps between the bone and the implant, and the surface design can be potentially employed to solve some o f the clinical issues associated with revision surgery.  82  CHAPTER 8: RECOMMENDATION FOR FUTURE WORK This  animal test provided very promising results  on the  enhancements  of  osteoconduction and bone ingrowth on porous tantalum by alendronate-immobilized calcium phosphate coating. It is a technique that has potential clinical applications. Before any clinical trial could be discussed, the following experiments need to be conducted for further confirmation o f the results.  1.  Systematic study on the effect o f alendronate loading amounts to optimize the drug dosage and to achieve the best effects on bone/implant fixation.  2. More in depth study on the mechanisms o f how alendronate immobilized calcium coating enhances osteoconductivity and new bone formation towards porous tantalum implants.  3. Design and launch the mechanical tests to collect data on the bone/implant fixation. Furthermore, confirm the enhancement o f bone/implant fixation on alendronate immobilized calcium phosphate coated implants.  4.  Systematic study o f new bone/implant fixation against varying postoperative ingrowth time.  83  REFERENCES 1. Just what is osteoporosis? Retrieved on Feb. 25 , 2007 from A women's perspective website:http://www.cumc.columbia.edu/dept/partnership/newsletters/volumel_Issuse 2/osteoporosis.html 2.  Y o u can be strong and vulnerable at the same time. Retrieved on Feb. 2 5 , 2007 from W e b M D website: http://www.webmd.com/solutions/osteoporosis-risks/strongvulnerable  3.  Hospital Morbidity Database, C I H I , 1994-1995 and 2004-2005.  th  4. Freeman M A et al. Early migration and late aseptic failure o f proximal femoral prostheses. J Bone Joint Surg B r 1994; 76(3): 432-438. 5. Maloney W J et al. Fixation, polyethylene wear, and pelvic osteolysis i n primary total hip replacement. C l i n Orthop R e l Res 1999; 369:157-164. 6. Duncan C P et al. Editorial-Antibiotic Depots. J Bone Joint Surg A m 1993; 75-B (3):349-350. 7. Meraw SJ et al. Qualitative analysis o f peripheral peri-implant bone and influence o f alendronate sodium on early bone regeneration. J Peridontol 1999;70:1228-1233. 8. Tengvall P et al. Surface immobilized bisphosphonate improves stainless-steel screw fixation in rats. Biomaterials 2004; 25: 2133-2138. 9. Yoshinari M et al. Bone response to calcium phosphate-coated and bisphosphonate immobilized titanium implants. Biomaterials 2002; 23: 2879-2885, 10. L i E C et al. Zoledronic acid: A new parenteral bisphosphonate. C l i n Ther 2003; 25: 2669-2708. 11. Hoffman A A et al. Comparative study o f human cancellous bone remodeling to titanium and hydroxapatite coated implants. J Arthroplasty 1993; 8:157-66. 12. Charnley J. Arthroplasty o f the hip - a new operation. Lancet 1961; 277(7187): 11291132. 13. Baier R E . Surface behaviour o f biomaterials: The theta surface for biocompatibility. J Mater Sci Mater M e d 2006; 7(11):1057-62. 14. B o b y n J D et al. Clinical validation o f a structural porous tantalum biomaterial for adult reconstruction. J Bone and Joint Surg. 2004; 86(A) supplement 2: 123-129.  84  15. Curry J D . The Mechanical Adaptations o f Bones. First edition; Princeton U n i v . Press. N J U S A 1984; 24-37 16. Weiner S et al. The Material Bone: Structure-Mechanical Function Relations. A n n u Rev Mater Sci 1998; 28: 271-298 17. Rho J et al. Mechanical properties and the hierarchical structure o f bone. M e d Eng Phys 1998; 20 (2): 9 2 - 102 18. Lowenstam H A et al. O n Biomineralization. Oxford Univ. Press, Oxford, U K 1989 19. M a n n S et al. Biomineralization: Structural question at all length scale. J struct B i o 1999; 126(3): 179-181. 20. Currey J D . What determines the bending strength o f compact bone? J E x p Biology 1999;202:2495-2503. 21. Weiner S et al. Lamellar bone: Structure-function relations. J Struct Biology 1999; 126:241-255 22. Ryland B E et al. Histologic Analysis o f Bone Healing. In: A n Y H , Martin K L , eds. Handbook o f Histology Methods for Bone and Cartilage, Humana Press, Totowa, N e w Jersey U S A 2003; chapter.28; 375-390. 23. Davies J E . Understanding peri-implant endosseous healing. J o f Dental E d u 2003; 67(8): 923-949 24. Davies J E . In vitro modeling o f the bone/implant interface. Anat Rec 1996;245:426445 25. Dimitriou R et al. Current concepts o f molecular aspects o f bone healing. Injury 2005; 36: 1392-1404. 26. Neuhoff D D V M et al. Anodic Plasma Chemical Treatment o f Titanium Schanz Screws Reduces P i n Loosening. J Orthop Trauma 2005; 19(8): 543-550. 27. Park J B et al. Metallic Biomaterials. In Park J B et al. Biomaterials: principles and applications. C R C Press, Boca Raton 2003; Chapter 1: 1-20 28. Muratoglu O K et al. Alternate bearing surfaces in hip replacement In: Sinha R K editor. H i p replacements: current trends and controversies. N e w Y o r k : Marcel Dekker, 2002. p. 1-46.  85  29. Firkins P J , et al. Quantitative analysis o f wear and wear debris from metal-on-metal hip prostheses tested in a physiological hip joint simulator. Biomed Mater Eng 2001; 11: 143-157. 30. Galea G et al. Review article: Clinical effectiveness o f processed and unprocessed bone. Transfusion Medicine 2005; 3: 165-174. 31. B o b y n J D et al. Characteristics o f bone ingrowth and interface mechanics o f a new porous tantalum biomaterial. J Bone Joint Surg 1999; 81B: 907-914. 32. Tsao A K et al. Biomechanical and clinical evaluations o f a porous tantalum implant for the treatment o f early-stage osteonecrosis. J o f Bone & Joint Surg A m 2005; 87: 22-27. 33. Matsuno H et al. Biocompatibility and osteogenesis o f refractory metal implants, titanium, hafnium, niobium, tantalum and rhenium. Biomaterials 2001; 22:1253-1262. 34. B o b y n J D et al. Tissue response to porous tantalum acetabular cups, a canine model. J of Arthroplasty 1999; 14 (3): 347-354. 35. B o b y n J D et al. Fundamental principles o f biologic fixation. In: Morrey BF,ed. Reconstructive surgery o f the joints. N e w York, etc: Churchill Livingstone 1996. p. 75-94, 36. Sabokbar A et al. Bisphosphonates in bone cement inhibit P M M A particle induced bone resorption. A n n Rheum D i s 1998; 57: 614-618. 37. Schwarz E M et al. Quantitative small-animal surrogate to evaluate drug efficacy in preventin wear debris-induced osteolysis. J Orthop Res 2000; 18: 849-855. 38. Total hip replacement: What it is. H o w it helps. Retrieved on Dec 21st, 2006 from Anderson orthopaedic research institute website: http://www.aori.org/thr/thrwhat.html 39. Moorhouse J. Waiting game. Retrived on Dec 2 0 , 2006 from Canadian health care manager website: http://www.chmonline.ca/issue/article.jsp=20020201_210080_9280 th  40. Horowitz S M et al. Phamacologic inhibition o f particulate-induced bone resorption. J Biomed Mater Res 1996; 31: 91-96. 41. Shanbhag A S et al. Inhibition o f wear debri mediated osteolysis in a canine total hip arthroplasty model. C l i n Orthop Relat Res 1997; 344: 33-43. 42. M c G r o r y et al.: Effect o f femoral offset on range o f motion and abductor muscle strength after total hip arthroplasty. J Bone Joint Surg B r 1995; 77-B(6): 865-9  86  43. Pennock J e t al. Morse type tapers. J Arthroplasty 2002; 17: 773-8 44. Rosenberg A G . Cementless acetabular components: the gold standard for socket revision. J Arthroplasty 2003; 18 (No.3 Suppl 1): 118-20. 45. Cohen R. A porous tantalum trabecular metal: basic science. A m J Orthop 2002; 31(4):216-7. 46. Sumner D R et al. Locally delivered r h B M P - 2 enhances bone ingrowth and gap healing in a canine model. J Orthop Res 2004; 22:58- 65. 47. Sumner D R et al. Enhancement o f bone ingrowth by transforming growth factor-beta. J o f Bone and Joint Surg A m . 1995; 7 7 - A (8): 1135-1147. 48. Sunmer D R et al. Additive enhancement o f implant fixation following combined treatment with R H T G F - P 2 and R H B M P - 2 in a canine model. J Bone Joint Surg 2006; 88-A (4): 806-817. 49. LeGeros R Z . Calcium phosphate materials in restorative dentistry: A review. A d v Dent Res 1988; 2(1): 164-180 50. Koutsoukos P et al. Crystallization o f calcium phosphates-constant composition study. J A m Chem Soc 1980; 102: 1553 -1557 51. Ducheyne P et al. The effect o f calcium phosphate ceramic composition and structure on in vitro behavior. 1. Dissolution. J biomed mater res 1993; 27: 25-34. 52. Shi S L et al. Reinforcement o f hydroxyapatite bioceramic by addition o f Ti3SiC2. J A m Ceram Soc 2006; 89: 743-745. 53. Tsui Y C et al. Plasma sprayed hydroxyapatite coatings on titanium substrates Part 1: Mechanical properties and residual stress levels. Biomaterials 1998; 19: 2015-2029 54. Sun L M et al. Material fundamentals and clinical performance o f plasma-sprayed hydroxyapatite coatings: A review. J o f Biomed Mater Res 2001; 58: 570-592 55. Gross K A et al. Amorphous phase formation in plasma sprayed hydroxyapatite coatings. J o f Biomed Mater Res 1998; 39: 407-414 56. C u i F Z et al. Preparation o f calcium phosphate coating on porous tantalum. J o f Mater Sci Letter 1998; 17: 925-930. 57. Compbell A A et al. Surface-induced mineralization: A new method for producing calcium phosphate coatings. J Biomed. Mater. Res. 1996; 32: 111-118.  87  58. Liu Q et al. The role of surface functional groups in calcium phosphate nucleation on titanium foil: a self-assembled monolayer technique. Biomaterials 2002; 23: 3103— 3111. 59. Duan K et al. Surface modifications of bone implants through wet chemistry. J. Mater. Chern. 2006; 16: 2309-2321. 60. Zhitomirsky I. Cathodic electrodeposition of ceramic and organoceramic materials. Fundamental aspects. Adv Colloid Interface Sci 2002; 97: 277-315. 61. Duan K et al. Electrochemical deposition and patterning of calcium phosphate bioceramic coatings. Ceramic Trans 2003; 47: 53-61. 62. Fleisch H . Bisphosphonates in bone disease: from the laboratory to the patient. 4 edition. San Diego: Academic Press, 2000.  th  63. Sabokbar A et al. Bisphosphonates in bone cement inhibit P M M A particle induced bone resorption. Ann Rheum Dis 1998; 57: 614-618. 64. Schwarz E M et al. Quantitative small-animal surrogate to evaluate drug efficacy in preventing wear debris-induced osteolysis. J Orthop Res 2000; 18: 849-855. 65. Horowitz S M et al. Phamacologic inhibition of particulate-induced bone resorption. J Biomed Mater Res 1996; 31: 91-96. 66. Lin JH et al. Physiological disposition of alendronate, a potent anti-osteolytic bisphosphonate, in laboratory animals. Drug Metabolism and Disposition 1991 19; 926-932 67. Bobyn JD et al. Zoledronic acid causes enhancement of bone growth into porous implants. J Bone Joint Surg Br 2005; 87B: 416-20. 68. Tanzer M et al. Bone augmentation around and within porous tantalum implants by local bisphosphonate elution. Clin Orthop Relat Res 2005; 441: 30-39. 69. Peter B et al. Local delivery of bisphosphonate from coated orthopaedic implants increases implants mechanical stability in osteoporotic rats. J Biomed Mater Res 2006; 76A: 133-143 70. Binderman I et al. Effectiveness of local delivery of alendronate in reducing alveolar bone loss fellowing periodontal surgery in rats. J Periodontal 2000; 71(8): 1236-40. 71. Delia Valle CJ et al. Revision of the acetabular component without cement after total hip arthroplasty. A concise follow-up, at fifteen to nineteen years, of a previous report. J Bone Joint Surg Am 2005; 87(8): 1795-800.  88  72. Zhang Y D et al. Interfacial frictional behavior: Cancellous bone cortical bone and a novel porous tantalum biomaterial. J Musculoskeletal Res 1999; 3: 245-251 73. K o l d S et al. Bone /compaction enhances fixation o f hydroxyapatite-coated implants in a canine gap model. J B i o m e d Mater Res Part B : A p p l Biomater 2005; 75B: 49-55. 74. Erben R G Bone-labeling technique. In: A n Y H , Martin K L , eds. Handbook o f Histology Methods for Bone and Cartilage, Humana Press, Totowa, N e w Jersey U S A 2003; chapter.5; 99-117. 75. Urist M R et al. Chemical reactivity o f mineralized tissue with oxytetracycline. A r c h o f Pathol 1963; 76: 484-496 76. Jansen J A Histological analysis o f bone-implant interface. In: A n Y H , Martin K L , eds. Handbook o f Histology Methods for Bone and Cartilage, Humana Press, Totowa, New.Jersey U S A 2003; chapter.26; 353-360. . ' 77. Pautke C et al. Polychrome labeling o f bone with seven different fluorochromes: Enhancing fluorochrome discrimination by spectral image analysis. Bone 2005; 37: 441 - 4 4 5 . 78. Frosch K H et al. Autologous osteoblasts enhance osseointegration o f porous titanium implants. J o f Orthop Res 2003; 21: 213 -223. 79. A l t m a n D G . Practical statistics for medical research. Chapman and H a l l / C R C Press, B o c a Raton, Florida U S A 1990 80. Paulson D S . Applied statistical designs for the researcher. Marcel Dekker, Inc., N e w York, N e w Y o r k U S A 2003 81. Medical terminology and drug database. Retrieved on March 2 , 2007 from St. Jude children's research hospital website: http://www.stjude.org/legal/0,262588_3170. htm n d  82. Deacon J. Student's t-test. Retrieved on Feb 2 5 , 2007 from the University o f Edinburgh website: http:// helios.bto.ed.ac.uk/bto/statistics/tress4a.html th  83. Dziewiatkowski D D . Isolation o f chondroitin sulphate- S from articular cartilage o f rats. J. B i o l . Chern. 1951; 189: 187-190 35  84. Bauer J H . Oral administration o f radioactive sulphate to measure extracellular fluid space in man. J. A p p l . Physiol. 1976; 40: 1976.  89  85. Bauer T W et al. Cutting and Grinding Methods for Hard-Tissue Histology. In: A n Y H , Martin K L , eds. Handbook o f Histology Methods for Bone and Cartilage, Humana Press, Totowa, N e w Jersey U S A 2003; chapter. 15; 2 3 3 - 242. 86. Bergstrom J D et al. Alendronate is a specific, nanomolar inhibitor o f farnesyl diphosphate synthase. A r c h Biochem Biophys 2000; 373(1):231-41 87. Wang J et al. Biomimetic and electrolytic calcium phosphate coatings on titanium alloy: physicochemical characteristics and cell attachment. Biomaterials 2004; 25: 583-92. 88. Fowler O B et al. Octacalcium phosphate. 3.Infrared and Raman vibrational spectra. Chern Mater 1993; 5: 1417-23. 89. Duan K . P h D thesis. The University o f British Columbia. Vancouver, B C . Canada 2007 90. Suchanek W et al. Processing and properties o f hydroxyapatite-based biomaterials for use as hard tissue replacement implants. J Mater Res 1998; 13: 94-117 91. W i l l m a n n G . Coating o f implants with hydroxyapatite material connections between bone and metal. A d v E n g Mater 1999; 1: 95-105 92. Chang Y L et al. Calcium and phosphate supplementation promotes bone cell mineralization: Implications for hydroxyapatite (HA)-enhanced bone formation. J Biomed Mater Res. 2000; 52: 270-278 93. Barrere F et al. Osteointegration o f biomimetic apatite coating applied onto dense and porous metal implants in femurs o f goats. J Biomed Mater Res 2003; 67B(1): 655-65. 94. Pidhorz L E et al. A quantitative study o f bone and soft tissues in cementless porouscoated acetabular components retrieved at autopsy. J Arthroplasty 1993; 8: 213-225. 95. K a n g J D et al. Ingrowth and formation o f bone in defects in an uncemented fibermetal total hip-replacement model in dogs. J Bone Joint Surg 73A:93- 105, 1991. 96. Barrack R L et al. Induction o f bone ingrowth from an acetabular defect to a porous surface with osteogenic protein-1. C l i n Orthop Relat Res 2003; 417: 41-9. 97. C o o k S D et al. The effect o f demineralized bone matrix gel on bone ingrowth and fixation o f porous implants. J Arthroplasty 17:402-408, 2002. 98. Tanzer M et al. Enhancement o f bone growth into porous intramedullary implants using non-invasive low intensity ultrasound. J Arthroplasty 2002; 19:195-199.  90  99. W i l k i n s o n J M et al. Effect o f pamidronate in preventing local bone loss after total hip arthroplasty: a randomized, double-blind, controlled trial. J Bone Miner Res 2001; 16: 556-64.  91  APPENDICES Appendix A: Student's t table (reprinted from ref 82 courtesy of Deacon J.)  Degrees of Freedom  Probability, p  0.1  0.05  0.01  0.001  1  6.31  12.71  63.66  636.62  2  2.92  4.30  9.93  31.60  3  2.35  3.18  5.84  12.92  4  2.13  2.78  4.60  8.61  5  2.02  2.57  4.03  6.87  6  1.94  2.45  3.71  5.96  7  1.89  2.37  3.50  5.41  8  1.86  2.31  3.36  5.04  9  1.83  2.26  3.25  4.78  10  1.81  2.23  3.17  4.59  11  1.80  2.20  3.11  4.44  12  1.78  2.18  3.06  4.32  13  1.77  2.16  3.01  4.22  14  1.76  2.14  2.98  4.14  15  1.75  2.13  2.95  4.07  16  1.75  2.12  2.92  4.02  17  1.74  2.11  2.90  3.97  18  1.73  2.10  2.88  3.92  19  1.73  2.09  2.86  3.88  20  1.72  2.09  2.85  3.85  21  1.72  2.08  2.83  3.82  22  1.72  2.07  2.82  3.79  23  1.71  2.07  2.82  3.77  24  1.71  2.06  2.80  3.75  25  1.71  2.06  2.79  3.73  26  1.71  2.06  2.78  3.71  92  27  1.70  2.05  2.77  3.69  28  1.70  2.05  2.76  3.67  29  1.70  2.05  2.76  3.66  30  1.70 •  2.04  2.75  3.65  60  1.67  2.00  2.66  3.46  120  1.66  1.98  2.62  3.37  00  1.65  1.96  2.58  3.29  93  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0078462/manifest

Comment

Related Items