Total hip replacements have been in use for over 30 years, and have shown great improvement from design to surgery since the first generation of implants were introduced. The greater need for hip replacements has led to the development of test methods that can be applied in the lab, which can predict the lifetime of a particular implant. To achieve this aim this study has sought to investigate key parameters, which may affect wear and its subsequent effect upon the production of particles for various material combinations and bearing geometries, under high (jogging) and low (walking) loads, with concurrent assessment of wear reduction, particle size and morphology. The clinical use of cross-linked polyethylene (XLPE) has been shown to reduce wear and thereby the onset of osteolysis in total hip arthroplasty. Hip simulator studies have suggested that while XLPE generates low wear under smooth counterface bearing surfaces, there appears to be an increased sensitivity to scratched femoral head conditions which can occur in the patient. However, these simulator studies have not combined damaged articular surfaces with a severe gait model, representing the worst-case scenario for high-risk, active patients. This hip simulator study has shown that the size distribution of wear particles generated in tests on 5 MRads crosslinked polyethylene can be influenced by the degree of patient activity. Fast jogging showed a greater influence on the number of sub-micron-sized wear particles (5-fold increase compared to walking) than on volumetric wear rate (26 mm3/106 cycles compared to 29 mm3/106 cycles). Fast jogging also did not generate the largest wear particles (>I 0p m) produced by normal walking. Roughening of the Co-Cr-Mo femoral heads created a 1700-fold increase in the numbers of sub-micron PE particles under fast jogging. The clinical significance of this result suggests that highly active patients will generate high numbers of bioactive PE wear particles within the accepted bioactive range, 0.2-10μm. Metal-on-metal (MOM) hip arthroplasty has also seen rapid growth worldwide. However, there remains concern over their long-term biocompatibility due to systemic ion release. Therefore, the aim of this current investigation was to test the hypothesis that larger diameter MOM bearings (greater than 40 mm) will generate smaller Co-Cr-Mo wear particles compared to a 28 mm size bearing, and reduce the total wear particle surface area, and to test the hypothesis that `severe' gait conditions will greatly increase the size of Co- Cr-Mo wear particles, thereby causing a sizable increase in wear particle surface area. Walking with a 28 mm bearing produced the largest wear rate of at 0.92 mm3/106 cycles, whereas the 40 mm and 56 mm bearings, generated lower wear rates of 0.39 mm3/106 cycles and 0.32 mm3/106 cycles respectively. Simulated fast jogging created a 3-fold increase in the number of elongated (needle) wear particles compared to normal walking, and generated a 20-fold increase in total wear particle surface area per year of use compared to normal walking. The clinical significance of this result suggests that highly active patients with MoM implants will exhibit greater ion release, although this may be minimised by using larger diameter bearings for active or younger patients.

The viscoelastic properties of soft lining materials are of enormous interest and importance as they clearly govern the biomechanical function of these materials. Forced vibration (Non-resonant) dynamic mechanical analysis was used to characterise the viscoelastic properties of a wide range of commercial (8 brands) as well as experimental soft liners (7 formulations). Forced vibration DMA is the ideal way of studying these materials as it can closely and accurately mimic physiological temperatures and frequencies under which they operate. The materials were studied as processed, and subsequently after simulated ageing for time periods up to twelve months. The water absorption and the subsequent changes in the viscoelastic behaviour of these materials are important as they spend a large part of their working lifetime immersed in solution. The results, where possible, were compared with the work of other researchers, and any sources of discrepancy examined. It was noted that the materials whose glass to rubber transition (Tg) was near their operating temperature (i.e. the acrylics) were particularly prone to any changes in the time and temperature domain of the experiment. A large variation was observed in the properties of the commercial materials available. At 37°C and 1 Hz a seven fold range was seen in the real part of the modulus (2.79-19.7 MPa), and a fifty fold range was observed in the loss tangent (0.029-1.52). Given this wide disparity it is surprising that a universally recommended range of properties does not exist, although there is some speculation in the literature. The study of the ageing behaviour of these materials in water showed that the traditional hardening of the acrylics has been overcome. A heat cured silicone material was seen to be virtually unaffected by time, while for a cold cured competitor the opposite was true. New and experimental materials showed a wide range of behaviour after immersion in water.

With time, the rate of symptomatic acetabular component loosening accelerates and overtakes that of the femoral component as the principal reason for the revision of total hip replacement. In the femur extensive study has shown that cement pressurisation and good preparation of the bone bed improves the survival rate, but acetabular fixation requires further investigation. Production of cement pressure in the acetabulum is anatomically difficult. Pressurisation with conventional and novel designs of cement pressurisers has been compared to manual techniques and component insertion. The pressurisers increased peak and mean pressures and pressure duration. Finite element modethng of cup insertion showed that flanges and higher insertion rates increased cement penetration into cancellous bone. Per-operatively, one design of pressuriser produced cement pressures comparable to those found in the laboratory. Structural finite element modelling of the natural hip indicated that the subehondral plate and the relatively dense cancellous bone supporting it distribute the joint contact force into the medial and lateral pelvic cortices. A perfectly bonded cemented polyethylene cup stiffened the acetabulum so that more load was transferred directly to the cortices at the acetabular rim, with consequent interface stress concentrations. However, complimentary experimental studies using a dynamic joint simulator and a servo-hydraulic materials testing machine suggested that perfect fixation between cement and bone at the rim was not possible, even under laboratory conditions. Debonding of the cement bone interface at the rim, where dense bone prevents cement interdigitation, allowed micromotion. Since the clinical mechanism of failure of the acetabular component appears to be progressive debonding, from rim to apex, of the cement-bone interface, these studies support the initiation of the failure mechanism by mechanical factors, which may then allow the ingress of wear debris. The experimental studies suggested that the use of pressurisers reduces the amount of micromotion and thus may improve the long term stability of the interface.

Spinal mobilisation or manipulation techniques are frequently used by physiotherapists in the treatment of musculoskeletal disorders. Despite the reliance on these techniques in clinical practice, there is little scientific evidence to substantiate their use. A standard mobilisation couch was instrumented to enable measurement of the forces applied to the trunk during mobilisation of the lumbar spine. Six load cells were incorporated into the couch frame and linked to a personal computer to facilitate data collection. The couch allowed the assessment of the magnitude of the mobilisation force, its direction and the variation in applied load over time. The system was found to be reliable and sensitive over the range of forces applied during mobilisation. The system was used to collect data from a sample of 30 experienced therapists to evaluate repeatability and reproducibility during the application of four grades of a posteroanterior mobilisation and an End Feel, on the third lumbar vertebra. Whilst some therapists demonstrated considerable variation in the forces applied both within one measurement session and over a two week period, others were found to be relatively consistent. The range of forces used by different therapists when performing the same technique was substantial ranging between 63 N and 347 N for a Grade IV mobilisation. A study was carried out involving 26 young healthy subjects, to determine the characteristics of a mobilisation force applied to an asymptomatic spine. A further study was undertaken involving a clinical sample of 16 patients, aged between 47- 64 years, to evaluate the effect of age related degenerative changes of the lumbar spine on the application of these techniques. The magnitude of the mobilisation force was found to be similar for the healthy and the patient groups with median forces of 175 N and 171 N during a Grade IV procedure, respectively. However, the forces applied to the patient group exhibited a statistically significantly smaller amplitude and higher frequency of oscillation than the healthy group for the same procedure (p < 0.01). Such measurements are essential for the assessment of the efficacy of these techniques in clinical practice.

Polypyrrole (PPy) is a conjugated polymer that displays special electronic properties including conductivity. It may be electrogenerated with the incorporation of any anionic species including negatively charged biological molecules such as proteins and polysaccharides. For this thesis, variously loaded-PPy films were prepared on gold sputter-coated coverslips. The growth and characteristics of epithelial cells, namely keratinocytes, were studied on these films by microscopy, biochemical assay, immunocytochemistry and electrochemical impedance spectroscopy. Keratinocyte viability was found to be PPy-load dependent. For chloride, polyvinyl sulphate, dermatan sulphate and collagen-loaded PPy films, polycarbonate and gold, keratinocyte viability, as assessed by the AlamarBlueTM assay, was respectively 47%, 60%, 88% and 23%, 75% and 61% of tissue culture polystyrene controls after 5 days. This was found to require a previously unreported polymer washing step prior to cell seeding due to the observed toxicity of untreated films. Keratinocytes stained positive for proliferation (PCNA), suprabasal differentiation (K10) and hyperproliferation (K16) markers although cell morphology was poor for organotypical cultures on dermatanloaded PPy compared with de-epidermalised dermis. Cell-induced impedance changes were detected in a three-electrode format over PPy modified electrodes. Results obtained showed the effects of cell density, cell type and monitoring frequencies. In particular, it was seen that lower cell densities could be detected on PPy compared to unmodified gold electrodes. Keratinocyte confluence as determined by impedimetric analysis was reached more rapidly on PPy than bare gold in agreement with AlamarBlueTM measurements. Electrical equivalent circuit analysis using parameters whose contributions may be directly mapped to intracellular and intercellular spaces, and membrane components suggested that the technique can be extended to cell morphology discrimination. This work shows that PPy biocomposites are attractive candidates for tissue engineering applications since they may incorporate biomolecules and are electrically addressable with the potential to both direct and report on cell activities.

Bioceramics have been developed for implants to repair damaged tissues of the human musculo-skeletal system. The clinical success of a bioceramic implant depends largely on the chemical response at the implant interface in addition to the sufficiency of the mechanical properties for the application. The present study combines the developments in the fields of bioceramic materials and laser surface micro structuring of materials. Bioceramic hydroxyapatite powders (HA, Formula: CaIO(PO4)6(OH)2) have been produced by emulsion technology and freeze-drying methods exhibiting BET specific surface areas >148 m2 /g and particle sizes <13 nrn prior to thermal treatment. The powder yield has been doubled using an increased reaction temperature of 25 *C from 17 *C, with a small increase (< 4nm) in the average particle size. HA discs that were >95.5 % dense have been achieved after isostatic pressing with pressure of 0.59 MPa and pressurelesss intering at 1200 *C for 2 hours. No chemical decomposition was detected using X-ray Diffraction Analysis (XRD). Methods of chemically enhanced laser-assisted etching have been developed to produce microstructural features on the surface of bioceramic HA discs that were 78.5 % dense (2482.95 kr , /M3 measured density). The use of 10 MPa SF6 at laser fluencies in the range of 14.50-15.20 W/M2 produced a columnar topography with individual structures featuring 10-20 pm height and 8-12 pm width as characterised by Scanning Electron Microscopy (SEM). Chemical characterisation by X-ray microdiffraction, Energy Dispsersive, X-ray analysis (SEM-EDS), Fourier Transformed Infrared spectroscopy (FTIR) and Raman spectroscopy (Raman) found the microtopography to be composed of fluorine-substituted HA (FHA). Alternatively the use of 80 MPa NH3 at laser fluencies in the range of 17.17-18.50 kj/m 2 produced an irregular surface of scattered porous hillocks that remained chemically unchanged in composition but exhibited four times as many surface hydroxyl groups. In both cases the mechanical and chemical stability of the bulk composition is maintained and the surface of increased surface area, in addition to the presence of concavities and pores is likely to be of enhanced osteoconductivity.

Silks have been described as biocompatible materials with a range of excellent mechanical properties, particularly a rare combination of strength and elasticity thought to arise from their unique molecular composition and ultrastructure. This project was aimed at studying the potential of natural silk fibres for tissue engineering. Species studied included degummed brins from the domestic silkworm Bombyx mori, the wild silkworm Antheraea pernyi and egg sac silk fibres from the spider Nephila edulis. A qualitative and quantitative description of the physical properties of the silks was carried out, where surface appearance, morphology and ultrastructure were investigated using a variety of microscopy techniques. The intrinsic fluorescence, density, linear density and cross section of silk fibres were also measured, facilitating a numerical estimation of their surface area per unit weight. Consequently, the tensile properties of the silks were measured, prior to and post cell culture preparation. The effect of autoclaving, washing and storage in growth medium was assessed through different tensile parameters. Upon culturing endothelial cells on the silk, a marked toxic effect was recorded and investigated, and different washing procedures to remove the toxic effect have been tested. Cell adhesion studies compared the ability of the different silks to support cell attachment for up to ten days. Also measured was as the effect of different treatments of the silk and scaffold design on rates of cell attachment. Finally, a new method of surface modification was developed and tested in order to functionalise natural silk fibres for endothelial guidance and angiogenesis. Results showed that all three silks shared a similar hierarchical ultrastructure of nanofibrils bundled into microfibrils, running parallel to the axis of the fibre. The silks also shared a lined, ridged topography where the fibroin monofilament was exposed or smooth topography when still coated in gum. However, tested silks differed in diameter, morphology, amino acid content and intrinsic fluorescence. Tensile tests showed that preparation for cell culture procedures, including sterilisation by autoclaving and water wetting had little or no effect on the mechanical properties, with the exception of medium incubation, which had a statistically significant effect on the mechanical properties of all tested silks. Cell growth studies showed that exposure of endothelial and myofibroblast cells to silk reduced their growth rates. The effect was mediated by iii both direct and indirect exposure to all tested silks, but most markedly by A. pernyi, which caused a severe cytostatic effect. One study showed that incubation of A.pernyi silk in medium supplemented with serum (but not non-supplemented medium) resulted in highly toxic growth medium. Among the different washing procedures devised to remove the toxicity, only lengthy enzymatic degumming was effective in reducing the toxic effect. Cell adhesion and growth studies indicated endothelial cells could attach and grow on the silk, but adhesion improved after the enzyme treatment. Scaffold design was also shown to have some effect on adhesion, with three-dimensional woven fabric proving a better scaffold than a random mesh of fibres. Finally, it was reported that the transglutaminase Factor XIII might be used to modify the surface of silk fibres with the biologically active factor L1Ig6. To conclude, studies presented in this described two poorly characterised silk species (Antheraea pernyi cocoon and Nephila edulis egg case) in term of their morphology, ultrastructure and fluorescence. They then showed that native, nonmodified silks supported cell attachment and growth provided they were treated to remove toxic coating and used in the form of a woven fabric rather then a loose mesh of fibres. Studies also presented a quantitative approach to the utilisation of silk in tissue engineering, and established wild silkworm silk Antheraea as a superior scaffold compared with egg case from the spider Nephile edulis or the domestic silkworm Bombyx mori. Finally, the novel method of silk modification using Factor XIII was reported as a potential route to further enhancing native silk fibres as cell scaffolds for specific applications. These studies present a unique approach, as they intentionally avoided harsh modifications of silk fibres before their use as scaffolds in order to preserve their excellent mechanical properties.

Materials used for medical devices are far from ideal for the body, so polymers are applied at the outermost region to counteract the body's natural defences. Component-based failures such as delamination and biocompatibility-based failures such as membrane fouling and degradation still remain a signifi cant challenge. This study focuses on the surface properties and modi cation of polyurethane and silicone rubber as coating material for amperometric sensing devices. E ffective synthesis of polyurethane, as well as surface modifi cation techniques performed on polymers already attached to sensing devices, are proposed. Phase inversion resulted in increased soft segment content on the surface (confi rmed by FTIR with a decreased C=O/C=C ratio). It is proposed that such an optimised polymer surface enhances the yield of further surface modi cation, such as hydroxylation (using potassium peroxodisulphate) and sulphonation (employing sodium hydride, triisobutylaluminium and 1,3 propane sultone). A novel method to generate an SO3-derivatised PU surface was proposed. Additionally, successful synthesis of silicone rubber for induced permeability of H2O2 was demonstrated. The physical and chemical properties of these modifi ed polymers were examined and evaluated via FTIR, SEM, TGA and contact angle measurements. The biocompatibility of modifi ed polymers was confi rmed with retarded protein adsorption; cytotoxicity testing showed that polymers were non-toxic to cells. Steady state amperometry on polymer modifi ed needle-type electrodes showed enhanced performance with surface modifi ed polymers to oxygen and H2O2, both of which are potential biological assay targets. Synthesised Prussian Blue (redox mediator) on platinum surfaces showed that the electrochemical response to H2O2 was increased threefold; and in combination with sulphonated polyurethane, interfering current responses could be successfully eliminated.

Polymer organosilicate nanocomposites have attracted significant attention over the last decade due to improved mechanical, thermal, and barrier properties. Several nanocomposite researchers have recognised potential for biomedical applications, however none have conducted biological investigations. In this project, the predicted ability of the organosilicate to enhance biostability, modulate the release of included drugs, and confer biofunctionality and control over the host response, were assessed as the three primary hypotheses. The studies were conducted with the objective being employment as urinary device biomaterials. Of prime importance was that no detrimental change in cytocompatibility was resultant. Biomedical thermoplastic elastomeric polyurethane organosilicate nanocomposites were prepared from poly(ether)urethane of 1000g/mol poly(tetramethylene oxide) polyol, 4,4??? diphenylmethane diisocyanate, and 1,4 butanediol chain extender chemistry, and various organosilicates with loadings from 1w% to 15w%, using a solution casting technique. Initially, partially exfoliated nanocomposites were produced using a commercially available organosilicate, Cloisite?? 30B. These nanocomposites displayed several advantageous properties, namely i) significant anti-bacterial activity, reducing S. epidermidis adherence after 24h to ~20% for a 15w% organosilicate loading, ii) enhanced biostability, with a 15w% organosilicate loading displaying only slight degradation after a 6 week subcutaneous in vivo ovine implantation, and iii) static modulation of model drug release as a factor of drug properties and organosilicate loading. The former was attributed to the Cloisite?? 30B quaternary ammonium compound, while the latter two were likely primarily barrier effect related and due to changes to poly(ether)urethane permeability. Electrostatic and chemical interactivity between drug and organosilicate was also implicated in the observed drug release modulation. Unfortunately, a lack of in vitro cytocompatibility and poor in vivo inflammatory response will limit in vivo use. Utilising bioinert 1-aminoundecanoic acid as an alternative organic modification, cytocompatible intercalated nanocomposites were produced thus likely allowing in vivo nanocomposite use and exploitation of the barrier effect related properties. However, these nanocomposites were not antibacterial. Variation of the organic modification and/or use of co-modification were viable means of modulating host response and biofunctionality, however nanoscale dispersion of co-modified silicate was poor. Use of nanocomposite technology was concluded beneficial to existing biomaterials, and specifically to biomaterial application as urinary catheters / stents.

Graduation date: 1999