Introduction

Post-arthroscopic glenohumeral chondrolysis (PAGCL) is a rare, but significant, complication of arthroscopic shoulder surgery that may lead to arthroplasty. Exact causal factors and pathways associated with the development of PAGCL are unknown however a number of patient factors and surgical factors have been implicated. Suture is one of these potential causal factors and currently little is known about the body's immune response to commonly used orthopaedic sutures. The aim of this project is to examine the biological response to 3 commonly used orthopaedic sutures (Ethibond, Fibrewire, and Orthocord) in a murine airpouch model. It was hypothesised that different sutures would elicit a different histological response and that suture wear-debris would induce an increased inflammatory reaction compared to intact suture.

Introduction

The need for regeneration and repair of bone presents itself in a variety of clinical situations. The current gold standard of treatment is autograft harvested from the iliac crest or local bone. Inherent disadvantages associated with the use of autogenous bone include limited supply, increased operating time and donor site morbidity. This study utilized a challenging model of posterolateral fusion to evaluate the in vivo response of an engineered collagen carrier combined with nano-structured hydroxyapatite (NanOss Bioactive 3D, Pioneer Surgical) compared to a collagen porous beta-tricalcium phosphate bone void filler (Vitoss BA, Orthovita).

The purpose of this study was to understand the effects of terminal sterilisation and residual calcium on human demineralised bone matrix (DBM) in ectopic bone formation in nude rat.

The intramuscular implantation of human DBM prepared by the Queensland Bone Bank (QBB) from four donors into eight male athymic rats was used to assess osteoinductivity. The DBM contained different levels of residual calcium and treated with or without gamma-irradiation at 11kGy. At 6 weeks post-implantation, calcium deposition was assessed by manual palpitation and radiological imaging. Tissue morphology and cellular interactions was analysed using various histological staining methods whilst protein expression of anabolic and catabolic biomarkers were examined through immunohistochemistry. All results were then analysed in qualitative, semi-quantitative and quantitative manners and tested for statistical significance.

Bone formation was observed in all specimens at the gross level. This was confirmed by histology which revealed bony capsules surrounded by soft tissue in the muscle pockets and differences in tissue components. On a cellular level, variations in osteoclast expression were found between the two groups as well as amongst individual donors through statistical analysis which resulted in an imbalance of the expression of anabolic and catabolic markers. Furthermore, a positive relationship between residual calcium and new bone formation in gamma irradiated DBM samples was found. To date, no studies have compared the effect of calcium in gamma irradiated DBM.

Our results suggest that gamma irradiation even at low doses and residual calcium may affect new bone formation. Taken together, this study stresses the importance of selecting ideal conditions for graft processing and the need to identify an optimal level of irradiation and remaining calcium levels that confers a balance between osteoinductivity and sterility.

To set up an osteosarcoma mouse model with spontaneous lung metastasis and to identify a marker of osteosarcoma metastasis and to inhibit the marker against the invasive ability of an osteosarcoma cell line.

A human osteosarcoma orthotopic mouse model was set up by injecting 143B human osteosarcoma cells into mouse tibia. Type I insulin-like growth factor receptor (IGF-1R) and its downstream signalling factors were measured in samples from the primary tumor and the lung secondaries by immunohistochemistry. Human Alu mRNA expression was tested using in situ hybridization assay. A Matrigel assay was used to assess cell invasion ability under the interference of a MEK/ERK pathway specific inhibitor, U0126.

All fifteen mice showed tumour mass at the left tibia and lung metastasis. Human Alu expression in the primary and secondary tumours confirmed human origin of the tumour cells. Total IGF-1R, MEK, Akt, p38 and phosphorylated MEK (p-MEK), but not p-Akt and p-p38, were positive in both local tumours and lung secondaries. Leiomyosarcoma controls expressed p-Akt and p-MEK, but not p-p38. The 143B cells treated with U0126 had significantly lower in vitro invasion ability compared with controls.

The IGF-1R-MEK signalling pathway, particularly Ras/Raf/MEK/ERK, may play an important role in osteosarcoma lung metastasis, and the targeting MEK/ERK by its specific inhibitor may have a potential use in the effective treatment of osteosarcoma.

Animal studies examining tendon-bone healing have demonstrated that the overall structure, composition, and organization of direct type entheses are not regenerated following repair. We examined the effect of Low-Intensity Pulsed Ultrasound (LIPUS) on tendon-bone healing. LIPUS may accelerate and augment the tendon-bone healing process through alteration of critical molecular expressions.

Eight skeletally mature wethers, randomly allocated to either control group (n=4) or LIPUS group (n=4), underwent rotator cuff surgery following injury to the infraspinatus tendon. All animals were sacrificed 28 days post surgery to allow examination of early effects of LIPUS. Humeral head – infraspinatus tendon constructs were harvested and processed for histology and immunohistochemical staining for BMP2, Smad4, VEGF and RUNX2. All the growth factors were semiquantitative evaluated. T-tests were used to examine differences which were considered significant at p < 0.05. Levene's Test (p < 0.05) was used to confirm variance homogeneity of the populations.

The surgery and LIPUS treatment were well tolerated by all animals. Placement of LIPUS sensor did not unsettle the animals. Histologic appearance at the tendon-bone interface in LIPUS treated group demonstrated general improvement in appearance compared to controls. Generally a thicker region of newly formed woven bone, morphologically resembling trabecular bone, was noted at the tendon-bone interface in the LIPUS-treated group compared to the controls. Structurally, treatment group also showed evidence of a mature interface between tendon and bone as indicated by alignment of collagen fibres as visualized under polarized light. Immunohistochemistry revealed an increase in the protein expression patterns of VEGF (p = 0.038), RUNX2 (p = 0.02) and Smad4 (p = 0.05) in the treatment group. There was no statistical difference found in the expression patterns of BMP2. VEGF was positively stained within osteoblasts in newly formed bone, endothelial cells and some fibroblasts at the interface and focally within fibroblasts around the newly formed vessels. Expression patterns of RUNX2 were similar to that of BMP-2; the staining was noted in active fibroblasts found at the interface as well as in osteoblast-like cells and osteoprogenitor cells. Immunostaining of Smad4 was present in all cell types at the healing interface.

The results of this study indicate that LIPUS may aid in tendon to bone healing process in patients who have undergone rotator cuff repair. This treatment may also be beneficial following other types of reconstructive surgeries involving the tendon-bone interface.

Fibrocartilaginous entheses are formed through endochondral ossification and characterized by four zones morphologically separated into tendon, uncalcified fibrocartilage, calcified fibrocartilage and bone [1]. These zones are not successfully regenerated following surgical repair. Demineralized Bone (DBM) presented at the tendon bone interface may improve healing between tendon and bone.

Fifty six female nude rats were randomly allocated into either a control reconstruction or treatment group (DBM at the tendon-bone healing site). A modified rodent model of anterior cruciate ligament reconstruction was adopted [2]. Animals were sacrificed at 2, 4 and 6 weeks following surgery. Four rats per group were prepared for histology at each time point while eight rats were culled for biomechanical testing at 4 and 6 week time points. ANOVA and post hoc tests were used to examine differences which were considered significant at p < 0.05.

The surgical procedure was well tolerated. Macroscopic dissection did not reveal any infection and all joint surfaces appeared normal. An intra-articular graft between the femur and tibia was present in all specimens. Mechanical differences were noted between groups. Peak loads were significantly higher in treatment group at 4 and 6 weeks (6.0 ± 3.6N and 9.1 ± 2.6 N, respectively) compared to controls (2.9 ± 1.9 N and 5.8 ± 2.7 N). No statistical differences were found in graft stiffness between the groups at 4 or 6 week time points. Histology showed an initial influx of inflammatory cells coupled with formation of a loose disorganized fibrovascular interface layer between tendon and bone in both groups. By the 6 weeks the interface layer in the DBM group fused into the newly formed bone to create a continuum between the tendon and bone, in an interdigitated fashion, containing Sharpy's like fibres. In the control group the continuum was less apparent with evidence of large areas of discontinuity between the two zones. A thicker region of newly formed woven bone with increased osteoblast activity along the bone tunnel was evident in the DBM group.

DBM has the potential to increase the quality of repair following surgical procedures involving reattachment of tendon to bone.

The biological properties of morselised bone allograft treated with either a supercritical fluid process or low-dose (15 kGy) gamma irradiation were compared using radiological, histological and immunohistological techniques. The aims were to investigate any differences in the biological properties of supercritical fluid treated allograft and low-dose gamma irradiated allograft in-vivo.

Rabbit allograft were cleaned of all soft tissue, cartilage and processed into ‘corticancellous crunch’ using a Noviomagus Bone Mill. Pooled samples were either gamma irradiated (15 kGy) or treated by NovaSterilis using super critical carbon dioxide. A well-reported tibial defect model in ten rabbits was used to examine the in vivo response of the different treatments at two and four weeks following surgery (n=5 per time point). Radiographic (x-ray, CT and micro CT), histology and immunohistochemistry was used to assess the in vivo response.

Radiographic results revealed an initial response to the gamma-irradiated samples compared to SCF. Histology confirmed this reaction to be inflammatory in nature at two weeks that continued at four weeks for the gamma irradiated samples. In contrast, the SCF treated sample demonstrated new bone formation while the inflammatory reaction was muted compared to the gamma irradiated samples. Four week x-rays and histology confirmed new bone formation in both groups while the lack of significant inflammatory response in the SCF group was noted.

Allograft sterilisation techniques do not result in the same initial response when evaluated in vivo. Removal of lipids and cellular debris following SCF treatment may influence the in vivo response. While both techniques can provide a sterile product, the in vivo response requires further investigation.

Posterolateral spinal fusion using autograft in adult rabbits has been reported by many groups using the Boden model. Age in general has an adverse effect on skeletal healing; although, its role in posterolateral fusion is not well understood. This study examined the influence of animal age on spinal fusion using a standard model and experimental endpoints. We hypothesised that fusion quality and quantity would be less with increasing age.

A single level posterolateral fusion between the fifth and sixth lumbar segments were performed in six-month and two-year-old New Zealand white rabbits (n=6 per group) using morcelized iliac crest autograft. All animals were sacrificed at 12 weeks following surgery. Posteroanterior Faxitron radiographs and CT scans were taken and DICOM data was analysed (MIMICS Version 12, Materialise, Belgium). Axial, sagittal, coronal and three-dimensional models were created to visualise the fusion masses. Bone mineral density (BMD) of the fusion mass was measured using a Lunar DPXL Dexa machine. An MTS Bionix testing machine was then used to assess peak load and stiffness. Sagittal and coronal plane histology was evaluated in a blinded fashion using H&E, Tetrachrome and Pentachrome stains. Assessment included overall bony response on and between the transverse processes. Radiographs and CT confirmed a more robust healing response in younger animals. Radiographic union rates decreased from 83% to 50% in the aged animals. A neo- cortex surrounding the fusion mass was observed in the younger group but absent in the aged animals. Fusion mass BMD and that of the vertebral body was decreased in the older animals (P<0.05). Tensile mechanical data revealed a 30% reduction in peak load (P=0.024) and 34% reduction in stiffness (P=0.073) in the two-year-old animals compared with the six-month-old animals. Histological evaluation demonstrated a reduction in overall biological activity in the two-year-old animals. This reduction in activity was observed in the more challenging intertransverse space as well as adjacent to the transverse processes and vertebral bodies at the decortication sites. Numerous sites of new bone formation was present in the middle of the fusion mass in the six-month-old animals while the bone graft in the two-year- old animals were less viable.

Skeletal healing is complex and mediated by both local and systemic factors. This study demonstrated that ageing leads to an impaired and delayed skeletal repair.

Where autograft is utilised, diminished graft osteoinductivity and reduced levels of growth factors and nutritional supply in the surrounding milieu explains our observations. The aged rabbit posterolateral spinal fusion model has not been previously described but would be a useful to evaluate new treatment modalities in a more challenging host environment.

Single level posterolateral spinal fusion in rabbits is the accepted preclinical model for evaluating bone graft substitutes or treatments to enhance/augment healing. This study aimed to improve preclinical testing by developing a multi-level unilateral fusion model that could be used as a screening tool prior to larger scale preclinical experiments.

A four level unilateral posterolateral fusion was performed in nine animals. The materials were randomly allocated and placed between the decorticated surfaces of the transverse processes and vertebral bodies. Animals were euthanised at three, six and 12 weeks. The materials were (1) 25 kGy y-irradiated rabbit allograft chips (RAC), (2) SCF RAC, (3) 60% tri-calcium phosphate, 40% hydroxyapatite formagraft (BiOstetic) (4) Autograft (1.5 cc morsellised to 1-2.5 mm granules). The autograft was harvested from the iliac crest using the L5-L6 incision. Endpoints included x-ray, CT, micro CT and histology.

The animals tolerated the surgery well. Radiographic data provided a useful method to differentiate between groups. Micro CT however was extremely valuable demonstrating new bone formation as early as three weeks across the groups. Gamma irradiated samples demonstrated an initial inflammatory reaction while the autograft, SCF allograft and synthetic TCP did not show this response. As expected, time was an important factor demonstrating the maturity in the fusions. These materials responded in a similar fashion in this model as observed in a single level fusion.

A unilateral multi-level fusion can be performed in rabbits to provide a useful screening for different materials. Gamma irradiated allograft has an initial inflammatory reaction that may be related to the presence of residual cellular material whereas SCF and synthetic materials do not.