Introduction: Cell adhesion to titanium alloy implants is important in osseointegration [1,2] and attachment of the soft tissues to skin penetrating implants e.g. external fixator pins and Intraosseous Transcutaneous Amputation Prostheses [3,4]. Cell adhesion can be assessed using cell area data and immunolocalisation of focal contact proteins e.g. vinculin; however no method of assessing biophysical attachment is performed routinely. Cell adhesion can be enhanced with adhesion proteins including fibronectin (Fn)[5]. We have previously shown that covalently binding Fn to titanium also increases cell adhesion, and produces a more robust protein coating [6]. However the strength of adhesion of cells to this coating has not been measured. Our hypothesis was that biophysical cell adhesion measured using novel radial flow apparatus would correlate with cell area and focal contact data and that covalently bound fibronectin substrates would increase cell adhesion compared with adsorbed and uncoated controls.

Method: Dermal fibroblasts were cultured for 1, 4, and 24 hours on 30mm and 10mm diameter polished titanium alloy discs (n = 6). Cells on 30mm discs were calcein stained and subjected to shear stress in a submerged, media filled, custom-made radial flow apparatus at 37¬C at 1.66ml/s for 15s. Cells were fixed in 10% formal saline and photographs were taken using a tangential light source. Fluorescent microscopy was performed at 2mm intervals along two perpendicular diameters. Using image analysis, the central cell free zone was measured and radial distance and shear stress calculated. Cells on 10mm discs were fixed, permeablised and vinculin stained (mouse vinculin antibody (1:200) 2hrs; FITC mouse antibody (1:100)1hr). Images were analyzed with a Zeiss microscope linked to image analysis software and the number of focal contacts were counted per cell area. The medians of the radial flow data were compared with data for cell area and focal contact production at the same time points using Spearman¡s regression correlation. This method was subsequently used to compare cell adhesion at one hour with adsorbed and covalently bound Fn substrates (10¥ìg/disc).

Results/Discussion: The shear strength of cells increased between 4 and 24hrs (p=0.002) on polished untreated control substrates. Attachment values (dynes/cm2) were 84.90 (73.98–97.19), 96.30 (91.66–100.89), and 136.69 (134.68–140.30) for 1, 4 and 24 hours respectively. At 1hr, covalently bound Fn (509.90 dynes/cm2 (490.55–528.49) significantly increased cell adhesion compared with adsorbed Fn(434.45 dynes/cm2(385.25–465.62)) and control substrates(p=0.002). There was significant correlation between shear stress and focal contacts/cell (1.00(p< 0.01)) and focal contacts/cell area (0.900(p=0.037)), but not cell area (0.600(p=0.285)).

Conclusion: Radial flow measurement is a useful direct method to quantify cell adhesion to orthopaedic implants and correlates well with other methods of measurement. Covalently bound Fn significantly increases biophysical cell attachment compared with adsorbed and uncoated controls.

Introduction: Aseptic loosening at the bone-implant interface of THA acetabular components is a significant cause of implant failure. This loosening has been attributed either to wear particle-induced osteolysis or to the effects of joint fluid-pressure. It may be possible to prevent the loosening of implants by improving fixation between the bone and implant, or promoting the growth of a biological bony seal, in order to prevent the influx of wear particles or pressurized joint fluid. Additionally in revision implants it is important to promote osseointegration in situations where bone stock may be limited. The hypothesis of this study was spraying autologous BMSCs in fibrin glue onto the surface of HA-coated acetabular components would increase bone formation around the implant and improve bone-implant contact.

Materials and Methods: Bone marrow was aspirated from the iliac crest of six goats, and BMSCs isolated and expanded in vitro. 10 x 10e6 BMSCs were suspended in reconstituted thrombin pre-operatively. A standard posterior approach was used. The acetabular shell was then coated with 2 ml of fibrin glue, with (n=6) or without 10 x 10e6 autologous BMSCs (n=6), and the acetabular component impacted into position. Antibiotic and analgesic prophylaxes were carried out. All animals were weight bearing within 48 hours post-operatively. Walking and ground reaction forces were assessed pre-operatively, as well as 6 and 12 weeks post-operatively. Results were expressed as a percentage of force transmitted through the right leg versus the left leg. After 12 weeks, the acetabulae were retrieved, and processed for histology. The percentage of new bone around the cups was measured within 5 radial zones, using image analysis. Bone-implant contact was also analysed between the new bone and implant surface. Mann Whitney U test was used to show statistical significance.

Results: New bone formation in Zone 5 showed a significant increase in the BMSC group (71.97±10.91%), when compared to the controls (23.85±15.13%, p=0.028). The other zones did not show a significant difference. Overall new bone growth in the BMSC group was 30% greater than the control group (71.42±8.97% and 54.22±16.56%, respectively, p=0.58). Bone-implant contact was significantly improved in the BMSC group (20.03±4.64%), in contrast to the control group (13.71±8.32%, p=0.027). With regards to the force plate analysis, there was no significant difference in loading between groups at both 6 weeks (Controls-79.74±3.63%, BMSCs-59.39±9.33%, p=0.086) and 12 weeks (Controls-86.0%±2.85%, BMSCs-62.33±5.12%, p=0.055).

Discussion and Conclusions: In this study, overall bone growth was greater when cups were treated with BMSCs. Bone-implant contact was significantly improved as well. This study has clinical applications, as using MSCs in fibrin glue promotes a bony seal in contact with the implant which may prevent the migration of particles, or joint fluid, decreasing the likelihood of aseptic loosening of THAs, and improving their longevity. Also, this technique may improve fixation in situations where bone stock is reduced.

Introduction: Finite element (FE) simulation of damage accumulation in the femoral cement mantle is widely used to predict failure of hip prostheses. It is often assumed that the stem-cement interface remains bonded, although debonding is thought to affect cement stress and damage. Rough stems may reduce subsidence, but have been reported to have a detrimental effect on implant survival. Other factors thought to influence cement damage include stem design and orientation and cement thickness. This study investigates the effect of cement mantle thickness and stem malpositioning on cement damage around a smooth, collared implant, and the extent to which this is affected by debonding of the stem-cement interface.

Method: Three FE meshes were built to represent proximal femora with Stanmore Hip prostheses implanted into a thick (2.5 mm) and a thin (1.0 mm) cement mantle, and another thin (1.0 mm) mantle with the implant tilted in varus to achieve a minimal thickness of 0.1 mm laterally. Each model consisted of 4304 eight-noded brick elements with frictional contact at the stem-cement interface. Two analyses were run for each model, in which the stem-cement interface was (a) fully bonded, and (b) fully debonded, with Coulomb frictional contact using a friction coefficient of 0.5. Standardised femur geometry and elastic properties were used. Creep and non-linear damage accumulation in the cement mantle under cyclic loading was modelled using subroutines developed by Stolk et al. (2003). Boundary conditions were applied representing a peak stair-climbing load.

Results: Bonded cases showed extensive cracking around the tip in all cases. Debonded cases had 4–8 times less cracking, which was much more focused at the tip; only the poorly-centralised mantle showed extensive damage elsewhere, in the very thin lateral region. When bonded, the thick mantle had least cracks and the poorly-centralised mantle had most; in the debonded cases, there was no major difference between thick, thin, and poorly-centralised mantles. For each cement mantle geometry, peak maximum principal cement stress was consistently lower in the debonded case than in the bonded case.

Discussion: Our results show greater, more widely distributed cracking in bonded than debonded cement mantles, in contrast with previous studies involving collarless implants. For a collared stem, calcar contact prevents subsidence, allowing cement stress relaxation. A possible explanation for our result is that debonding enhances the stress relaxation process, reducing and redistributing interfacial and shear stresses; thus reducing damage rates. In contrast, a debonded collarless stem subsides continuously, sustaining high cement stress levels and damage rates. These results may explain the disappointing clinical performance of some rough-surfaced prostheses. Our results suggest that bonding might increase both cement damage and its sensitivity to cement thickness. Similar results for all debonded cement mantles indicate that cement thickness may be less critical than previously thought for smooth, collared prostheses. Bonding should not be assumed in FE studies of smooth stems which clinically are likely to debond; cement damage simulation should be extended to incorporate the debonding process.

Introduction: Incorporation of Silicon into the HA structure enhances the bioactivity of Hydroxyapatite (HA). Silicon substituted calcium phosphate (SiCaP/SiHA) has been introduced as an osteoconductive material for bone formation. However, the osseoinductive capacity of this biomaterial has not been assessed. A previous study by Hing et al shows that bioactivity of stoichiometric hydroxyapatite bone substitute materials is enhanced by increasing the level of porosity within the implant struts [1].

The aim of this study was to test the hypothesis that SiCaP bone graft results in superior osseoinduction compared to stoichiometric HA and osseoinduction enhancement using high microporosity materials.

Methods: Implantation of 32 bone graft plugs (16 granular and 16 blocks) with 3 different strut porosities: 20% SiHA, 35% SiHA, 10% SiHA and 20% HA, all with matched 80% total porosity supplied by ApaTech Ltd into the paraspinalis muscle of 4 sheep for 12 weeks. HA and %SiHA locations were randomized at implant sites.

Following euthanasia at 12 weeks histomorphometry was carried out to calculate Percentage of bone, soft tissue and implant area and Percentage of the amount of bone in contact with the calcium phosphate surface (% Bone attachment). Further evaluation of Calcium, Phosphate and Silicon levels within the implants and surrounding bone was carried out by Scanning Electron Microscopy (SEM) and EDAX.

Results: Bone formation was observed within the pores of both granules and blocks of SiCaP and HA implants. Greater bone formation and attachment was detected in scaffolds with higher strut porosity (SiHA35) compared to implants of the same chemical composition but lower strut porosity (SiHA10, SiHA20. More bone formation and contact was observed in SiHA implants (SiHA20) compared to matched porosity HA implants where the amount of bone formed was minimal. Uniform distribution of Silicon (Si) was visible within the SiHA scaffold struts according to EDAX results. Greater quantities of Si existed in newly formed bone as compared to soft tissue adjacent to the SiHA implants. Silicon was not detected in either soft or hard tissues adjacent to HA implants.

Conclusion: Both microporous HA and SiCaP promote bone ingrowth, as ectopic bone formation was observed in all four groups of synthetic materials. Matched porosity SiCaP is more osseoinductive than HA. Increasing strut porosity results in promotion of osseoinductivity. High strut porosity (> 10%) block environment contributes to greater osseoinductive behaviour. In conclusion we report that presence of silicon and the strut porosity influence the osseoinductive capacity of calcium phosphate bone substitute biomaterials.

Introduction There are 1 million cases of major skeletal defects :that occur worldwide each year that lead to significant morbidity and disability and currently require bone grafting as the main mode of treatment. Limitations of bone-grafting include donor site morbidity, reduced osseoinductivity and risk of pathogen transmission to the host. There is considerable interest in finding ways of differentiating mesenchymal stem cells down the osteoblastic lineage to form bone tissue. We hypothesized that there is an optimum strain that promotes differentiation of mesenchymal stem cells into osteoblasts.

Methods: A bioreactor was developed that was capable of applying tensional forces across a culture strip in a graduated manner within a range of 1-4373me. Mesenchymal stem cells were grown on these strips and subjected to cyclical tensile strain at 1Hz. Cell morphology using Scanning Electron Microscopy, mineralization using specialized stains and expression of core binding factor1 (Cbfa1) was studied at various strain levels.

Results: Scanning Electron Microscopy revealed classic osteoblastic cells in the regions subjected to tensile force, especially in the region where average strain was 1312me. X-ray microanalysis revealed calcium deposits on the strip, indicating osteoblastic differentiation. Cbfa1 expression was greatest in the region with an average strain 1312 me followed by a region on the strip subjected to just fluid shear without any tension. Cbfa1 expression was significantly greater in cells subjected to tensile forces than unstrained controls at all levels of strain tested (p< 0.05). Cbfa1 expression was further enhanced significantly by the addition of osteogenic factors (p< 0.05). Significantly greater mineralization (p< 0.05) occurred in the regions subject to tension with the greatest being in the region with an average strain of 1312 me.

Conclusions: Mechanical tensile forces especially in the range of up to 2173me promote differentiation of Mesenchymal Stem Cells into osteoblasts and encourage expression of the Cbfa1 gene. Tensile strain also promotes mineralization. Chemical factors in form of osteogenic media accelerate the differentiation of MSCs and encourages earlier production of osteoblast specific markers. Fluid shear appears to have a beneficial effect in stimulating differentiation into the osteoblast phenotype and, combined with tensile strain, may offer an even greater osteogenic stimulus.

We treated 98 patients with peri-acetabular tumours by resection and reconstruction with a custom-made pelvic endoprosthesis. The overall survival of the patients was 67% at five years, 54% at ten years and 51% at 30 years. One or more complications occurred in 58.1% of patients (54), of which infection was the most common, affecting 30% (28 patients). The rate of local recurrence was 31% (29 patients) after a mean follow-up of 71 months (11 to 147). Dislocation occurred in 20% of patients (19). Before 1996 the rate was 40.5% (17 patients) but this was reduced to 3.9% (two patients) with the introduction of a larger femoral head. There were six cases of palsy of the femoral nerve with recovery in only two. Revision or excision arthroplasty was performed in 23.7% of patients (22), principally for uncontrolled infection or aseptic loosening. Higher rates of death, infection and revision occurred in men.

This method of treatment is still associated with high morbidity. Patients should be carefully selected and informed of this pre-operatively.

Introduction & Aims: A new femoral component for hip arthroplasty has been designed for a younger patient population. The design makes use of a higher femoral cut, which conserves bone stock, increasing options for future revision surgery. It uses the existing load bearing properties of the proximal femur, and therefore distributes load more evenly. The stem is longer than that of a resurfacing, so will be easier to insert at the correct orientation, minimising failure rates in inexperienced hands. The cross-sectional dimensions have been designed to produce torsional stability. The collar maximises the loading of the calcar, reducing stress resorption. The surface is hydroxyapatite coated and porous, which will produce a long-term biological fixation.

This project assessed the long-term stability of this design at different orientations, by measuring the change in surface strain distribution following its insertion.

Methods: Ten composite bones were coated in a Photoelastic material, positioned at a simplified single leg stance, and loaded at 2.3 KN. The surface strain was measured at one-centimetre intervals down the medial cortex. Then the prostheses were inserted into the bone at 135°, 145° and 125° to the femoral shaft, and the surface strains reread.

Results: The results were compared with an FEA model, and analysed statistically using the Wilcox signed rank test. The prosthesis inserted at 135° produced no significant difference in surface strain distribution compared with the intact bone.

Conclusions: This study suggests this stem design will be stable in the long term following insertion, and there were no areas of excessively high or low strain.

We previously demonstrated that cartilaginous tissue was induced on a reamed acetabular articulation in an ovine hemiarthroplasty model with three different femoral head sizes. At maximum loading during stance phase, the acetabular peak stresses immediately after reaming could reach approximately 80 MPa under direct implant-bone contact with in-vitro measurements.

We aimed to establish finite element (FE) models of the ovine hip hemiarthroplasty which examine stress distribution on the reamed acetabula by three head sizes. We hypothesized that the stress distribution did not differ between different sizes when the joint is congruent and that the peak stresses in the acetabulum immediately after reaming occurred in the dorsal acetabulum.

Three two-dimensional FE models of ovine hip hemi-arthroplasty were built; each comprised a head component, 25, 28, and 32 mm in diameter, and an acetabular component. The acetabular geometry was acquired from an ovine acetabular histological section. The head was moved to partly intersect with the acetabulum representing the reaming procedure and a congruent contact was confirmed. Cortical bone and cancellous bone were modelled as linear elastic, with moduli of 20 and 1.2 GPa, respectively. Variable moduli were also assessed. The finest mesh for each model consisted of over 100,000 four-node quadrilateral elements. Loading conditions were chosen to represent peak hip joint force developed during the stance phase. Stress distribution in the acetabular area in contact with the head was plotted against the articulating arc length.

The results confirmed that the stress distribution between different prosthetic head sizes in a reamed hemiarthroplasty model did not change when the joint was congruent. The peak compressive stresses occurred in the dorsal acetabulum with the 32 mm model being the highest at approximately 69 MPa, the 28 mm model at 63 MPa, and the 25 mm model at 54 MPa. An increase in the cancellous modulus and a decrease in the cortical modulus increased the peak stresses in the dorsal acetabulum.

This presents an indicative study into the effect of prosthetic femoral head sizes on the stress distribution in the acetabulum. The idealized 2-D models showed reasonable agreement when compared quantitatively with the in vitro study.

Introduction: Many mini compression screws are now available for fixation in procedures such as metatarsal osteotomies or arthrodeses of the foot.

The aim of the current study is to compare the compression forces achieved by mini compression screws on cortical and cancellous bone models.

Material and Methods: The screws that were tested are listed in the table below. The compression forces were tested by inserting a pressures load measurement cell between longitudinally-split sheep tibia as a cortical bone model and longitudinally split retrieved femoral heads as a cancellous bone model.

Results: The Headed AO 3.5 mm cortical screw gave the best compression force and the Bold was the weakest, both in cortical and cancellous bone. The relative compression forces of the other tested screws were different between cortical and cancellous bone. Compression with the headless screws was lost as soon as the screw penetrated through the cortex in the cortrical bone model.

Conclusions: The indications for using headless self-tapping screws should be reserved for fixation of cancellous bone or of metatarsal or Akin osteotomies where compression is not required for union. When compression is important, such as in MPJ, tarso-metatarsal or talonavicular arthrodeses, Headed AO 3.5 mm or 2.7 mm cortical or 4 mm cancellous screws, which give better compression, should be used.

Background: As the understanding of bone repair mechanics has advanced the integrity of the bone pin interface has emerged as a key factor in determining the success of external fracture fixation. The benefits of using pins coated with Hydroxyapatite (HA) are well documented however the thickness of the conventional plasma spray coating precludes its use for modification of the surface of fine features in implants. Consequently new electro-chemical techniques for pre-coating implants with a ‘biomimetic’ HA layer using simulated body fluids (SBF) have been pioneered. In this study we test the hypothesis that varying the technique for deposition of HA by electrolysis of SBF alters the morphology of the HA surface which will modify the level of osseointegration. Method: Three alternative methods of HA coating the Barerre, Redepenning and Kumar techniques were compared. Tantalum coated stainless steel pins were coated then used to stabilise a mid-diaphyseal osteotomy in three sheep using an orthofix fixator for a period of ten weeks. Insertion and extraction torques were measured to calculate the pin performance index (PPI). Sections of the bones were then examined using scanning electron microscopy to determine the percentage of bone in contact with the pin surface and the percentage of new bone formation. Results: The different coating protocols resulted in different HA crystal morphologies. The extraction torque exceeded the insertion torque for both the Barerre and Redepenning methods and their PPI exceeds that of plasma spray coatings. The Redepenning technique was shown to perform sig-nificantly better than both the Barerre (p=0,001) and Kumar (p=0,001) techniques with 49.4% of the pin surface in contact with bone. These results were mirrored on analysis of new bone formation with the Redepen-ning technique showing 70.2% of new bone formation compared to the Barerre (55.4%) and Kumar (53.8%) methods. Conclusion: These results indicate that the Redepenning technique is the most effective for creating a bio mimetic HA coating in terms of bonding to bone and promoting new bone formation. This technique holds significant advantages over the conventional plasma spray technique for example the coating thickness can be easily controlled and additional proteins such as bone morphogenic proteins and antibiotics can be incorporated. It may therefore represent a new era in the use of HA coating.

Aims: The main purpose of this study was to evaluate hemiplevis replacements in patients with periacetabular tumours in terms of the functional results and implant survivorship of the prosthesis. The second aim was to evaluate the complications and how they might be prevented in the future.

Methods: Case notes, hospital databases and radiographs were reviewed in 51 patients treated in a 16 year period with a custom made hemipelvis replacement (Stanmore Implants Worldwide). There were 13 deaths and 4 were lost to follow up. Of the remaining 34 patients there were 18 males and 16 females with a mean age of 48.6 (range 14 – 71).

Results: At a mean follow-up of 78 months (range 7 – 204), 70% of patients were still alive and the most common diagnosis was chondrosarcoma (17 cases). There were 2 cases of prosthesis failure and 22 of the 34 patients had one or more complications (12 cases of infection (8 deep and 4 superficial), 4 nerve palsies, 2 dislocations and 7 others). 13 patients had to have a second operation. The mean MSTS score was 63.8% and the TESS score was 59.8%

Conclusion: Infection was the most common complication and was the most significant reason for reoperation. This method of reconstruction is associated with a high morbidity rate and should be performed only at specialist centres, but the functional and oncological outcomes are satisfactory.

Introduction: The current practice of impaction allograft to fill large defects in revision total hip replacements is sometimes useful but clinical results are inconsistent. Other studies have shown that addition of mesenchymal stem cells (MSC) in blocks of hydroxyapatite (HA) scaffold can enhance new bone formation in a critical sized defect. However, no study has been conducted on combined MSCs with morselised allograft and HA granules. It is hypothesized that impaction of allograft or HA granules seeded with MSCs or osteoprogenitors will enhance new bone formation compared with the groups without MSCs.

Materials and Methods: Six sheep were used for the study. Each sheep received 8 scaffolds which were embedded in both paraspinal muscles. Groups were: 1) 3.5g allograft, 2) 3.5g allograft with MSCs, 3) 3.5g allograft with osteoblasts; 4) 3.5g of 50:50 allograft/ HA, 5) 3.5g of 50:50 allograft/HA with MSCs, 6) 3.5g of 50:50 allograft/HA with osteoblasts; 7) a block of HA, 8) a block of HA with MSCs. The experimental scaffolds were seeded with either 10x106 MSCs/ml or 10x106 MSC-derived osteoprogenitors/ml, in 3ml autologous plasma. Grafts were impacted twenty times at 3KN. At eight weeks, samples were sectioned for histology analysis. Areas of new bone formation were measured as percentage to total available spaces. ANOVA was used for statistical analysis.

Results: Addition of MSCs increased new bone formation in allograft (4.98%), allograft/HA (5.15%) and HA block (7.09%) compared with their controls at 2.24%, 1.96% and 1.96% respectively. Statistical study showed significant increase in 50:50 allograft/HA with MSCs compared with 50:50 allograft/HA only (p=0.046) and 50:50 allograft/HA with osteoprogenitors (p=0.028). No difference was found in allograft groups. For the HA block groups, addition of MSCs showed a significant new bone increase compared to the control (p=0.028).

Conclusion: Addition of MSCs to the allograft and HA granules will enhance new bone formation after impaction which can be used for revision total hip replacements, especially when allograft and HA is mixed. However, addition of osteoprogenitors has not achieved the similar results. This study encourages a further clinical investigation of impaction tissue-engineered graft to repair bone defects in revision total joint replacements.

Introduction: Intraosseous Transcutaneous Amputation Prostheses (ITAP) could overcome the problems associated with conventional stump-socket prostheses for amputees (pressure sores, pain, infections and unnatural gait), by attaching the external prosthesis directly to the skeleton via a skin penetrating abutment. Despite this, the skin breach introduces a potential route for infection. For success, a biological seal at the skin-ITAP interface is essential.

The protein Laminin-5 (L-5) is a ‘biological glue’, which is integral to epitheial cell adhesion. Covalently bonding L-5 to the ITAP titanium alloy (Ti6Al4V), may enhance the strength of the skin-ITAP interface.

Silanisation, a chemical technique that covalently bonds proteins to metals, could be used to bond L-5 to Ti6Al4V. We have assessed the characteristics L-5 silanised Ti6Al4V as a potential substrate for ITAP.

Method: To determine the maximum quantity of L-5 that could be silanised to Ti6Al4V, and its relative stability when soaked in foetal calf serum (FCS) over time; polished Ti6Al4V discs were silanised by immersing in aminopropyltriethoxysilane followed by glutaraldehyde. Radiolabelled rat laminin-5-I125 was then added. Discs were immersed in FCS for 4 days (37 C) and analysed at 24 hour intervals in a liquid scintillation counter. Un-silanised discs were used as controls.

Results: L-5 was successfully covalently bound to Ti6Al4V. 10ng, 100ng, 250ng and 500ng droplets yielded significantly more silanised L-5 (p< 0.05), but no difference was observed between 750ng and 1000ng. Percentage L-5 covalently bound ranged from 33% and 65%.

A small decrease in bound L-5 occurred after 24 hours of FCS soaking (p< 0.05), but subsequent to this no significant reduction was observed for 4 days (p< 0.05). Controls showed a significantly larger reduction after 24 hours (p< 0.05).

Conclusion: Covalently bonding L-5 to Ti6Al4V by silanisation can be achieved with predictable results. Large enough quantities can be immobilised to influ-ence cellular function. L-5 silanised to Ti6Al4V remains stable in vitro over time and is not removed. Following the study of cellular interactions with silanised L-5, a stable skin seal may be achieved at the transcutaneous portion of the ITAP.

Aim: To study the remodelling of cortical bone around the passive growing component of an expandable endoprosthesis.

Introduction: When inserting the passive component of an expandable prosthesis a polyethylene sleeve is commonly used. The sleeve migrates towards the lateral cortex and causes a cortical reaction and hence the use of the sleeve has been discarded recently. This study quantifies the amount of cortical reaction and degree of cortical drift in patients that had sleeves and those that did not.

Material And Methods: We reviewed X-rays and case notes of all patients that had an expandable endoprosthesis in a 20 year period. The thickness of medial and lateral cortices of the tibial diaphysis was measured at 6 months and on the last follow up radiograph. The distance from the edge of the sleeve (or prosthesis) from the cortical edge was also compared. Retrieved components also had their histology reviewed.

Results: The sleeve shifted laterally on average by 2mm (range 0.5 – 3mm) and touched the cortex. This was associated with an increase in lateral cortical thickness by 2.27mm (range 1 – 3mm). When the sleeve was used the prosthesis was inserted in the mid-line. When the sleeve was not used the tibial component tended to be inserted in valgus.

Conclusions: The presence of a sleeve is associated with a cortical reaction and the sleeve tends to migrate laterally. The clinical implications of this and the evolution of the design will be discussed in the meeting.

Introduction: Since 1975, 6 types of extendable endo-prostheses have been developed at Biomedical Engineering, UCL, and Stanmore Implants Worldwide in conjunction with the surgeons at this centre.

Aims: To establish whether developments in design have had the desired effect of improving both implant survivorship and functional outcome.

Methods: This was a retrospective study using case notes, hospital databases and a radiological review, combined with contemporary functional outcome assessments (MSTS, TESS, SF36).

Results: 161 consecutive prostheses in 138 paediatric patients, between the years of 1983 – 2005, were implanted for primary bone tumours. Mean age was 10.3 (3 – 18), 81 were males and 57 females. There were 136 primary procedures and 25 revisions.

6 prostheses that used ball bearings to achieve length (designed in 1981) were implanted, 3 (50 %) were revised due to mechanical failure.

19 prostheses that utilised external C-washes (1998) to achieve length were implanted, 6 (32 %) were revised, half of these for mechanical failure.

Of the 98 minimally invasive prostheses (1992) that utilized an Allan key and screw-jack mechanism to lengthen, 14 (14 %) were revised, half of these for infection.

17 non-invasive extendable endoprostheses (2001) that are lengthened by electromagnetic coupling have been implanted so far. There has only been 1 (6 %) revision. This was due to full extension being reached.

Conclusions: Design improvements in growing endoprostheses since 1983 have led to improved survivorship. Initially this led to a reduction in mechanical failure and latterly to a reduction in infection, as indications for revision.

Key Words: Bone tumour, children, endoprosthesis, survivorship.

Bone grafts are frequently used to augment bone healing. Autologous bone graft is the gold standard for osteogenesis but is limited by availability and donor site morbidity. The processing required to lower the immunogenicity of allograft also reduces the osteogeneic properties. Bone marrow contains mesenchymal stem cells (MSCs) which differentiate into osteoblasts, forming bone. Our study examined the use of bone marrow to enhance the osteogenic properties of allograft.

Bioactive proteins within allogenic bone graft stimulate marrow-derived MSCs to differentiate into osteoblasts, thereby increasing the osteogenic nature of the graft.

After informed consent, bone marrow aspirates were taken from five patients during orthopaedic operations. Freeze-dried ethylene oxide treated allograft, from a number of donors, was obtained from the bone bank. MSCs isolated from each marrow aspirate were grown on eight samples of test allograft. Further allograft was heated to 70°C to denature the osteogenic proteins and MSCs from each aspirate were grown on 8 samples, as a negative control. Osteoblastic differentiation of MSCs cultured on the types of allograft was compared.

Scanning electron microscopy confirmed that MSCs covered the allograft after 14 days. Transmission electron microscopy showed that cells on the test allograft were characteristic of osteoblasts and produced collagen extracellular matrix. The levels of osteoblastic proteins, ALP, osteopontin and Type I pro-collagen, produced by cells on test allograft were significantly greater compared with heat-treated control (P< 0.005), after days 7 and 14.

Our study showed that marrow-isolated MSCs could be successfully cultured on allograft. As the levels of osteoblastic proteins increased significantly when MSCs were grown on allograft, osteogenic proteins within allograft caused MSCs to change into osteoblasts. This confirms that autologous marrow MSCs could be grown on allograft to increase its osteogenic prior to grafting, resulting in increased rate of bony healing.

The purpose of this study was to determine whether the low rate of mechanical loosening of the SMILES rotating hinge distal femoral endoprosthesis relates to the hydroxyapatite (HA)-coated, grooved collar of the femoral component.

A database was used to identify two groups of cases of primary distal femoral replacement with a custom-designed and manufactured SMILES endoprosthesis at our unit; those with the collared femoral component (“collar group”), and those without a collar (“non-collargroup”). From these two groups, patients were pair-matched for age and length of bone resection. A retrospective review of serial biplanar pairs of radiographs of each patient, assessing radiolucent lines and extracortical bone pedicle.

11 matched pairs were identified, (14 females, 8 males), with a mean age of 36 years, (range 16–66). The pathology was primary bone tumour in 20 cases, (17 malignant, 3 benign), and metastatic disease in 2 cases. Mean length of follow-up was 85 months, (range 27–122). Radiolucent line score (RLS) progression over time was significantly lower in the collar group, (0.01 vs 0.73, p=0.001) (fig. 1 & 2), as was the mean final RLS, (2.72 vs 7.81, p=0.02). Mean RLS per radiographic quadrant was 0.56 in cases in which a bony pedicle was ingrown onto the prosthesis, (exclusively in the collared-group), 2.41 in cases in which the pedicle was not ingrown, (most prevalent in the non-collared group), and 1.02 in those cases without any pedicle formation, (ANOVA analysis, p=0.0002).

This study demonstrates that the HA-coated, grooved collar significantly reduces the progression of radiolucent lines, and consequently the overall RLS, explaining the reduced rate of mechanical loosening of the collared endoprosthesis. A bony pedicle that does not incorporate onto the prosthesis surface may be associated with an increase in radiographic loosening.

Endoprosthetic reconstruction as a form of limb salvage in the management of malignant disease is common. We present our experience with custom-made distal femoral replacement as a form of limb salvage in the absence of malignancy.

49 cases of distal femoral replacement were identified using the unit database. There were 18 males and 31 females, with a mean age at operation of 62.3 years (range 26–86). There were 29 cases of failed total knee replacements, 8 cases of fracture associated with bone loss, non-union or deformity, 7 periprosthetic fractures, 3 aneurysmal bone cysts, and one case each of avascular necrosis and Gorham’s disease. Clinical and radiographic review of all available patients, including a functional assessment with the Musculoskeletal Tumour Society (MSTS) Score and Toronto Extremity Salvage (TES) score was undertaken.

The average follow-up was 5.4 years (range 1 to 29 years). Three types of endoprosthesis were used, rotating hinge, fixed hinge and arthrodesis prosthesis. One patient required amputation at 2 months following post-operative wound infection with methicillin-resistant staphylococcus aureus and subsequent inability to provide adequate soft tissue coverage. There was one revision at 16 months for deep prosthetic infection. 21 patients were available for functional follow-up. The mean MSTS score was 63.7 (range 16.0–86.7) and the mean TES score was 59.4 (range 9.4–87.5).

Custom-made distal femoral replacements have an established role in limb-salvage surgery for malignant disease. This series demonstrates the reconstructive capability of custom prostheses in non-malignant disease, where deformity causes functional embarrassment or when massive bone loss would normally lead to amputation.

Interventional MRI provides a novel non-invasive method of in-vivo weight-bearing analysis of the talo-calcaneal joint. Six healthy males (mean 28.8 years) underwent static right foot weight bearing MRI imaging at 0o, 15o inversion, and eversion. Using known radiological markers the motion of the talus and calcaneum were analysed.

The calcaneum externally rotates, plantar-flexes and angulates into varus. The talus shows greater plantarflexion with similar varus angulation, with variable axial rotation. Relative talo-calcaneal motion thus involves, 6o relative talar internal rotation, 3.2o flexion and no motion in the frontal plane. Concurrently the talus moves laterally on the calcaneum, by 6.5mm, with variable translations in other planes.

The calcaneum plantar-flexes, undergoes valgus angulation, and shows variable rotation in the axial plane. The talus plantar-flexes less, externally rotates, and shifts into varus. Relative motion in the axial and saggital plane reverses rotations seen during inversion. The 8o of relative valgus talo-calcaneal angulation is achieved through considerable varus angulation of the talus, in a direction opposite to the input motion. This phenomenon has not been previously reported. From coronal MRI data, comparative talo-calcaneal motion in inversion is prevented by high bony congruity, whereas during eversion, the taut posterior tibio-talar ligament appears to prevent talar valgus angulation.

We have demonstrated that Interventional MRI scanning is a valuable tool in analysing the weight-bearing motion of the talo-calcaneal joint, whilst approaching the diagnostic accuracy of stereophotogammetry. We have also demonstrated consistent unexpected talar motion in the frontal plane. Talo-calcaneal motion is highly complex involving simultaneous rotation and translation, and hence calculations of instantaneous axes of rotation cannot effectively describe talo-calca-neal motion. We would suggest that relating individual and relative motion of the talus / calcaneum better describes subtalar kinematics.

The purpose of this study was to evaluate the results of custom-made endoprosthetic reconstruction for both tumourous and non-tumourous conditions around the elbow joint.

28 consecutive cases of endoprosthetic elbow reconstruction, performed between 1989 and 2003, were identified using the unit database. There were 12 males and 16 females, with a mean age 48.9 years, (range 14–84). There were 16 cases of malignant tumour (10 primary, 6 secondary), 3 cases of benign tumour (one each of giant cell tumour, osteoblastoma and pigmented villonodular synovitis) and 10 non-oncological cases, (5 cases of fracture, 3 failed total elbow replacements and 1 infected synostosis). A distal humeral prosthesis was used in 16 patients, distal humeral and proximal ulna in 10 patients, proximal ulna with a humeral component in one patient and proximal radial replacement in one patient. Clinical and radiographic review of all available patients, including a functional assessment with The Toronto Extremity Salvage score (TESS) was undertaken.

28 consecutive cases of endoprosthetic elbow reconstruction, performed between 1989 and 2003, were identified using the unit database. There were 12 males and 16 females, with a mean age 48.9 years, (range 14–84). There were 16 cases of malignant tumour (10 primary, 6 secondary), 3 cases of benign tumour (one each of giant cell tumour, osteoblastoma and pigmented villonodular synovitis) and 10 non-oncological cases, (5 cases of fracture, 3 failed total elbow replacements and 1 infected synostosis). A distal humeral prosthesis was used in 16 patients, distal humeral and proximal ulna in 10 patients, proximal ulna with a humeral component in one patient and proximal radial replacement in one patient. Clinical and radiographic review of all available patients, including a functional assessment with The Toronto Extremity Salvage score (TESS) was undertaken.

Endoprosthetic reconstruction around the elbow joint is effective in a wide range of pathologies, allowing in most cases a reasonable level of function, even following two-stage revision for infection.