Abstract

Introduction: Revision arthroplasty using the impaction grafting technique is an increasingly popular technique. The lost bone stock is replaced rather than substituted with ever increasing amounts of metal. There have been many advances in the understanding of this technique in recent years. It has recently been shown that washing of the graft improves the biomechanical strength, bony ingrowth and biocompatibility of morsellised allograft bone. The aim of this study was to identify the most efficacious method of washing morsellised allograft in the operating room.

Methods: Fresh frozen femoral heads identified for research purposes were randomly divided into four groups. Group A was washed using a serial dilution technique at 37°C, group B serial dilution at 60 °C, group C pulsed lavage wash at 37°C and group D pulsed lavage wash at 60 °C. Three-gram samples of each femoral head were taken prior to washing. The heads were then washed according to the grouping. The unwashed and washed morsellised bone was then assessed for the presence of protein, haemoglobin, DNA and lipid using spectrophotometric analysis techniques. The removal of these marrow components from the femoral heads in each group was assessed as a proportion of the original content.

Results: All washing techniques removed a large proportion of the marrow load of the femoral heads. The most efficacious technique was washing by serial dilution (removal rate 67.5–87.4%). Washing by pulsed lavage removed less marrow (28.4–80.9%). Protein showed the highest removal rate (80.9–86.4%) and haemoglobin the lowest (28.4–67.5%). Removal of lipid was most affected by temperature with a larger proportion removed at 60°C as compared to 37°C.

Conclusion: We conclude that the most efficient method of washing morsellised allograft bone for impaction grafting is by serial dilution with fluid at 60°C. This information should improve the results of impaction grafting, as it is known that cleaner bone results in improved biomechanical stability, enhanced biological compatibility and accelerated bony ingrowth.

Impaction grafting with morsellised allograft is becoming the treatment of choice for revision arthroplasty, especially in the younger patient. The optimum treatment of the graft prior to impaction has not been determined. Some groups wash the graft prior to impaction and others do not. Washing of graft has been shown to enhance bone ingrowth in an animal model, however the reasons for this remain unclear. The aim of this study was to identify any underlying cellular cytotoxicty of fresh frozen allograft bone before and after washing.

Paired samples of washed and unwashed morcellised FFH allograft were taken during revision hip arthroplasties. Washing was performed by 4 consecutive rinses in 300ml warmed saline, the bone being filtered between each exchange of saline. Contact cytotoxicity assays involved culture of cell lines in direct contact with bone samples. Quantitative cytotoxicity assays utilised culture media conditioned with the bone samples and subsequent assessment of cell metabolism and viability using both dimethylthiazol (MTT) and neutral red (NR) assays. Assays were performed with human osteoblastic (MG63) and fibroblastic (HSF) cell lines.

Nine pairs of samples were analysed. Contact assays demonstrated a clear zone of cellular inhibition around the unwashed bone samples. Quantitative assays were performed in triplicate for each cell type and both MTT and NR assays giving 108 paired assay results. 88.9% of pairs (92/108) showed cytotoxicity in the unwashed sample. No washed samples demonstrated cytotoxicity. When grouped by assay and cell type, analysis of means showed statistically significant differences between washed and unwashed samples in MG63-NR (p=0.0025), HSF-NR (p=0.0004) and MG63-MTT (p=0.008). The difference observed in the HSF-MTT assays did not reach statistical significance (p=0.06).

Unwashed FFH allograft can be cytotoxic to human osteoblastic and fibroblastic cell lines in vitro. This suggests that allograft should be washed prior to impaction in order to optimise the biological compatibility.

Introduction: The success of impaction-grafting depends on mechanical stability and adequate bony incorporation of the graft. Full incorporation of this type of graft has been demonstrated histologically and depends on many factors including the biological activity of the graft. Bone morphogenic proteins (BMPs) are known to play a central role in bone formation and their presence reflects the biological activity of a graft material. The aim of this study was to determine the activity of fresh frozen femoral head (FFH) grafts by analysing BMP-7 release after milling and during strain imposed by the impaction process.

Methods: 10mm cancellous bone cubes were cut from 5 samples of FFH. The cubes were washed, centrifuged and washed again to remove the marrow contents. Specimens from each femoral head were allocated to five groups and subjected to strains of 0%, 20%, 40%, 60% and 80% with a material testing machine. The cubes were washed again and the wash fluid analysed for bmp-7 activity using a commercially available elisa kit. Additionally, samples of bone were taken after standard milling of FFH, washed and the fluid analysed for bmp-7 activity.

Results: bmp-7 activity was found to be present in all groups. Release of bmp-7 was found to increase with increasing strain in a linear relationship. At 80% strain the mean concentration of bmp-7 released (2.2 ng/g bone) was approximately double that released at 20% strain.

Discussion: activity of bmp-7 in FFH has not previously been demonstrated. This study shows that the freezing and storage of femoral heads allows some maintainance of biological activity. Furthermore we have shown that bmp-7 may be released from FFH cancellous bone in proportion to the strain applied to the bone. This may go some way to explaining the full bony incorporation often seen after impaction-grafting.

Fresh frozen femoral head (FFH) allograft is commonly used in impaction grafting for revision hip arthroplasty and long term success has been demonstrated by some groups. The optimum treatment of the graft prior to impaction has not yet been determined. Some groups wash the graft prior to impaction and others do not. Washing of the graft has been shown to improve bone ingrowth in a bone chamber animal model however the reasons for this remain unclear. The aim of this study was to identify any underlying cellular cytotoxicty of fresh frozen allograft bone before and after washing.

Samples of morcellised FFH allograft were taken during revision hip arthroplasties prior to impaction grafting. Paired samples, taken before and after washing were taken from each case. Washing was performed by 4 consecutive washes in 300ml warmed saline, the bone being filtered between each exchange of saline. Cytotox-icity was assessed for all samples using both contact and extract assays. Contact assays involved culture of cell lines in direct contact with bone samples. Extract assays utilised culture media conditioned with bone samples and subsequent quantitative assessment of cell metabolism and viability using both dimethylthiazol (MTT) and neutral red (NR) assays. All assays were performed using both human osteoblastic (MG63) and fibroblastic (HSF) cell lines.

Nine pairs of samples were analysed for cytotoxicity using both cell lines. Contact assays demonstrated a clear zone of cellular inhibition around the unwashed bone samples. Extract assays were performed in triplicate for each cell type and both MTT and NR assays giving 108 paired assay results. 88.9% of pairs (92/108) showed cytotoxicity in the unwashed sample. No washed samples demonstrated cytotoxicity. When grouped by assay and cell type, analysis of means showed statistically significant differences between washed and unwashed samples in MG63-NR (p=0.0025), HSF-NR (p=0.0004) and MG63-MTT (p=0.008). The difference observed in the HSF-MTT assays did not reach statistical signifi-cance (p=0.06).

In conclusion, we have shown that unwashed FFH allograft can be cytotoxic to human osteoblastic and fibroblastic cell lines in vitro. This suggests that allograft should be washed prior to impaction in order to optimise the biological compatibility.

Bone morphogenic proteins (BMPs) are members of the transforming growth factor beta (TGF-beta) family and play a central role in bone formation. These morpho-gens are known to be present in bone matrix however the characteristics of their release during the grafting process has not previously been defined. The aim of this study was to determine the release BMP-7 (osteogenic protein; OP-1) from cancellous allograft that occurs during impaction grafting for revision hip arthroplasty.

Forty, 10mm cubes of cancellous bone were accurately cut from the central region of 7 fresh frozen femoral heads. The cubes were centrifuged and washed to remove the marrow contents. The cubes were then individually washed and the fluid assayed for BMP-7 activity using a commercially available enzyme linked immuno-sorbent assay kit (Raybiotech Inc.). The cubes were then divided into 4 groups with samples from each femoral head in each group. Each group was subjected to strain of either 20%, 40%, 60% or 80% using a material testing machine. The cubes were then individually washed again and the wash fluid analysed for BMP-7 activity.

BMP-7 activity was found to be present in all groups. Release of BMP-7 was found to increase with increasing strain. At 80% strain the mean concentration of BMP-7 released (830 pg/g) was 58% greater than that released at 60% strain (527 pg/g), 150% greater than the concentration at 40% strain (333 pg/g) and 476% greater than at 20% strain (144 pg/g). The differences between release at 80% and 40% strain and between 80% and 20% strain were statistically significant (p=0.036, p=0.002).

Activity of BMP-7 in fresh frozen cancellous allograft bone has not previously been demonstrated. This study shows that the freezing and storage of femoral heads allows some maintenance of biological activity. Furthermore we have shown that BMP-7 may be released in proportion to the strain applied to the bone. This confirms that the process of impaction of bone morsels during revision hip arthroplasty may release BMPs that could aid in the incorporation and remodelling of the allograft.

In order to investigate the osteoinductive properties of allograft used in impaction grafting and the effect of strain during impaction on these properties, we designed an in vitro experiment to measure strain-related release of bone morphogenetic protein-7 (BMP-7) from fresh-frozen femoral head allograft. A total of 40 10 mm cubes of cancellous bone were cut from ten samples of fresh-frozen femoral head. The marrow was removed from the cubes and the baseline concentrations of BMP-7 were measured. Specimens from each femoral head were allocated to four groups and subjected to different compressive strains with a material testing machine, after which BMP-7 activity was reassessed. It was present in all groups. There was a linear increase of 102.1 pg/g (95% confidence interval 68.6 to 135.6) BMP-7 for each 10% increase in strain. At 80% strain the mean concentration of BMP-7 released (830.3 pg/g bone) was approximately four times that released at 20% strain. Activity of BMP-7 in fresh-frozen allograft has not previously been demonstrated. This study shows that the freezing and storage of femoral heads allows some maintenance of biological activity, and that impaction grafting provides a source of osteoinductive bone for remodelling.

We have shown that BMP-7 is released from fresh-frozen femoral head cancellous bone in proportion to the strain applied to the bone. This suggests that the impaction process itself may contribute to the biological process of remodelling and bony incorporation.