Hydroxyapatite-coated standard anatomical and customised femoral stems are designed to transmit load to the metaphyseal part of the proximal femur in order to avoid stress shielding and to reduce resorption of bone. In a randomised in vitro study, we compared the changes in the pattern of cortical strain after the insertion of hydroxyapatite-coated standard anatomical and customised stems in 12 pairs of human cadaver femora. A hip simulator reproduced the physiological loads on the proximal femur in single-leg stance and stair-climbing. The cortical strains were measured before and after the insertion of the stems. Significantly higher strain shielding was seen in Gruen zones 7, 6, 5, 3 and 2 after the insertion of the anatomical stem compared with the customised stem. For the anatomical stem, the hoop strains on the femur also indicated that the load was transferred to the cortical bone at the lower metaphyseal or upper diaphyseal part of the proximal femur. The customised stem induced a strain pattern more similar to that of the intact femur than the standard, anatomical stem

We analysed revised Mathys isoelastic polyacetal femoral stems with stainless-steel heads and polyethylene acetabular cups from eight patients in order to differentiate various types of particle of wear debris. Loosening of isoelastic femoral stems is associated with the formation of polyacetal wear particles as well as those of polyethylene and metal. All three types of particle were isolated simultaneously by tissue digestion followed by sucrose gradient centrifugation. Polyacetal particles were either elongated, ranging from 10 to 150 μm in size, or shred-like and up to 100 μm in size. Polyethylene particles were elongated or granules, and were typically submicron or micronsized. Polyacetal and polyethylene polymer particles were differentiated by the presence of BaSO. 4. , which is added as a radiopaque agent to polyacetal but not to polyethylene. This was easily detectable by back-scattered SEM analysis and verified by energy dispersive x-ray analysis. Two types of foreign-body giant cell (FBGC) were recognised in the histological specimens. Extremely large FBGCs with irregular polygonal particles showing an uneven, spotty birefringence in polarised light were ascribed to polyacetal debris. Smaller FBGCs with slender elongated particles shining uniformly brightly in polarisation were related to polyethylene. Mononucleated histiocytes containing both types of particle were also present. Our findings offer a better understanding of the processes involved in the loosening of polyacetal stems and indicate why the idea of ‘isoelasticity’ proved to be unsuccessful in clinical practice

We have examined 26 retrieved, failed titanium-alloy femoral stems. The clinical details, radiological appearances and the histology of the surrounding soft tissues in each patient were also investigated. The stems were predominantly of the flanged design and had a characteristic pattern of wear. A review of the radiographs showed a series of changes, progressive with time. The first was lateral debonding with subsidence of the stem. This was followed by calcar resorption and fragmentation or fracture of the cement. Finally, osteolysis was seen, starting with a radiolucency at the cement-bone interface and progressing to endosteal cavitation. Three histological appearances were noted: granulomatous, necrobiotic and necrotic. We suggest that an unknown factor, possibly related to the design of the stem, caused it to move early. After this, micromovement at the cement-stem interface led to the generation of particulate debris and fracture of the cement. A soft-tissue reaction to the debris resulted in osteolysis and failure of fixation of the prostheses

CT and advanced computer-aided design techniques offer the means for designing customised femoral stems. Our aim was to determine the Hounsfield (HU) value of the bone at the corticocancellous interface, as part of the criteria for the design algorithm. We obtained transverse CT images from eight human cadaver femora. The proximal femoral canal was rasped until contact with dense cortical bone was achieved. The femora were cut into several sections corresponding to the slice positions of the CT images. After obtaining a computerised image of the anatomical sections using a scanner, the inner cortical contour was outlined and transferred to the corresponding CT image. The pixels beneath this contour represent the CT density of the bone remaining after surgical rasping. Contours were generated automatically at nine HU levels from 300 to 1100 and the mean distance between the transferred contour and each of the HU-generated contours was computed. The contour generated along the 600-HU pixels was closest to the inner cortical contour of the rasped femur and therefore 600 HU seem to be the CT density of the corticocancellous interface in the proximal part of cadaver femora. Generally, femoral bone with a CT density beyond 600 HU is not removable by conventional reamers. Thus, we recommend the 600 HU threshold as one of several criteria for the design of custom femoral implants from CT data

Our aim was to determine the precision of the measurements of bone mineral density (BMD) by dual-energy x-ray absorptiometry in the proximal femur before and after implantation of an uncemented implant, with particular regard to the significance of retro- and prospective studies. We examined 60 patients to determine the difference in preoperative BMD between osteoarthritic and healthy hips. The results showed a preoperative BMD of the affected hip which was lower by a mean of 4% and by a maximum of 9% compared with the opposite side. In addition, measurements were made in the operated hip before and at ten days after operation to determine the effect of the implantation of an uncemented custom-made femoral stem. The mean increase in the BMD was 8% and the maximum was 24%. Previous retrospective studies have reported a marked loss of BMD on the operated side. The precision of double measurements using a special foot jig showed a modified coefficient of variation of 0.6% for the non-operated side in 15 patients and of 0.6% for the operated femur in 20 patients. The effect of rotation on the precision of the measurements after implantation of an uncemented femoral stem was determined in ten explanted femora and for the operated side in ten patients at 10° rotation and in 20 patients at 30° rotation. Rotation within 30° influenced the precision in studies in vivo and in vitro by a mean of 3% and in single cases in up to 60%. Precise prediction of the degree of loss of BMD is thus only possible in prospective cross-sectional measurements, since the effect of the difference in preoperative BMD, as well as the apparent increase in BMD after implantation of an uncemented stem, is not known from retrospective studies. The DEXA method is a reliable procedure for determining periprosthetic BMD when positioning and rotation are strictly controlled

We have compared the changes in the pattern of the principal strains in the proximal femur after insertion of eight uncemented anatomical stems and eight customised stems in human cadaver femora. During testing we aimed to reproduce the physiological loads on the proximal femur and to simulate single-leg stance and stair-climbing. The strains in the intact femora were measured and there were no significant differences in principal tensile and compressive strains in the left and right femora of each pair. The two types of femoral stem were then inserted randomly into the left or right femora and the cortical strains were again measured. Both induced significant stress shielding in the proximal part of the metaphysis, but the deviation from the physiological strains was most pronounced after insertion of the anatomical stems. The principal compressive strain at the calcar was reduced by 90% for the anatomical stems and 67% for the customised stems. Medially, at the level of the lesser trochanter, the corresponding figures were 59% and 21%. The anatomical stems induced more stress concentration on the anterior aspect of the femur than did the customised stems. They also increased the hoop strains in the proximomedial femur. Our study shows a consistently more physiological pattern of strain in the proximal femur after insertion of customised stems compared with standard, anatomical stems

We studied the effect of the surface finish of the stem on the transfer of load in the proximal femur in a sheep model of cemented hip arthroplasty. Strain-gauge analysis and corresponding finite-element (FE) analysis were performed to assess the effect of friction and creep at the cement-stem interface.

No difference was seen between the matt and polished stems. FE analysis showed that the effects of cement creep and friction at the stem-cement interface on femoral strain were small compared with the effect of inserting a cemented stem.

The role of vacuum mixing on the reduction of porosity and on the clinical performance of cemented total hip replacements remains uncertain. We have used paired femoral constructs prepared with either hand-mixed or vacuum-mixed cement in a cadaver model which simulated intra-operative conditions during cementing of the femoral component. After the cement had cured, the distribution of its porosity was determined, as was the strength of the cement-stem and cement-bone interfaces.

The overall fraction of the pore area was similar for both hand-mixed and vacuum-mixed cement (hand 6%; vacuum 5.7%; paired t-test, p = 0.187). The linear pore fractions at the interfaces were also similar for the two techniques. The pore number-density was much higher for the hand-mixed cement (paired t-test, p = 0.0013). The strength of the cement-stem interface was greater with the hand-mixed cement (paired t-test, p = 0.0005), while the strength of the cement-bone interface was not affected by the conditions of mixing (paired t-test, p = 0.275). The reduction in porosity with vacuum mixing did not affect the porosity of the mantle, but the distribution of the porosity can be affected by the technique of mixing used.

The use of impaction bone grafting during revision arthroplasty of the hip in the presence of cortical defects has a high risk of post-operative fracture. Our laboratory study addressed the effect of extramedullary augmentation and length of femoral stem on the initial stability of the prosthesis and the risk of fracture. Cortical defects in plastic femora were repaired using either surgical mesh without extramedullary augmentation, mesh with a strut graft or mesh with a plate. After bone impaction, standard or long-stem Exeter prostheses were inserted, which were tested by cyclical loading while measuring defect strain and migration of the stem. Compared with standard stems without extramedullary augmentation, defect strains were 31% lower with longer stems, 43% lower with a plate and 50% lower with a strut graft. Combining extramedullary augmentation with a long stem showed little additional benefit (p = 0.67). The type of repair did not affect the initial stability. Our results support the use of impaction bone grafting and extramedullary augmentation of diaphyseal defects after mesh containment

It has been suggested that the endurance of cemented femoral reconstructions in total hip arthroplasty is affected by the creep of acrylic cement, but it is not known to what extent cement creeps under loading conditions in vivo, or how this affects load transfer. We have simulated the long-term creep properties of acrylic cement in finite-element models of femoral stem constructs and analysed their effects. We investigated whether subsidence rates measured in vivo could be explained by creep of acrylic cement, and if polished, unbonded, stems accommodated creep better than bonded stems. Our findings showed that polished prostheses subsided only about 50 μm as a result of cement creep. The long-term prosthetic subsidence rates caused by creep of acrylic cement are therefore very small and do not explain the excessive migration rates which have sometimes been reported. Cement creep did, however, relax cement stresses and create a more favourable stress distribution at the interfaces. These trends were found around both the bonded and unbonded stems. Our results did not confirm that polished, unbonded, stems accommodated creep better than bonded stems in terms of cement and interface stress patterns

We revised 24 consecutive hips with loosening of the femoral stem using impaction allograft and a cemented stem with an unpolished proximal surface. Repeated radiostereometric examinations for up to two years showed a slow rate of subsidence with a mean of 0.32 mm (−2.0 to +0.31). Fifteen cases followed for a further year showed the same mean subsidence after three years, indicating stabilisation. A tendency to retroversion of the stems was noted between the operation and the last follow-up. Retroversion was also recorded when displacement of the stem was studied in ten of the patients after two years. Repeated determination of bone mineral density showed an initial loss after six months, followed by recovery to the postoperative level at two years. Defects in the cement mantle and malalignment of the stem were often noted on postoperative radiographs, but did not correlate with the degrees of migration or displacement. After one year, increasing frequency of trabecular remodelling or resorption of the graft was observed in the greater trochanter and distal to the tip of the stem. Cortical repair was noted distally and medially (Gruen regions 3, 5 and 6). Migration of the stems was the lowest reported to date, which we attribute to the improved grafting technique and to the hardness of the graft

We investigated the implant-bone interface around one design of femoral stem, proximally coated with either a plasma-sprayed porous coating (plain porous) or a hydroxyapatite porous coating (porous HA), or which had been grit-blasted (Interlok). Of 165 patients implanted with a Bimetric hip hemiarthroplasty (Biomet, Bridgend, UK) specimens were retrieved from 58 at post-mortem. We estimated ingrowth and attachment of bone to the surface of the implant in 21 of these, eight plain porous, seven porous HA and six Interlok, using image analysis and light morphometric techniques. The amount of HA coating was also quantified. There was significantly more ingrowth (p = 0.012) and attachment of bone (p > 0.05) to the porous HA surface (mean bone ingrowth 29.093 ± 2.019%; mean bone attachment 37.287 ± 2.489%) than to the plain porous surface (mean bone ingrowth 21.762 ± 2.068%; mean bone attachment 18.9411 ± 1.971%). There was no significant difference in attachment between the plain porous and Interlok surfaces. Bone grew more evenly over the surface of the HA coating whereas on the porous surface, bone ingrowth and attachment occurred more on the distal and medial parts of the coated surface. No significant differences in the volume of HA were found with the passage of time. This study shows that HA coating increases the amount of ingrowth and attachment of bone and leads to a more even distribution of bone over the surface of the implant. This may have implications in reducing stress shielding and limiting osteolysis induced by wear particles

Objectives

Infection of implants is a major problem in elective and trauma surgery. Heating is an effective way to reduce the bacterial load in food preparation, and studies on hyperthermia treatment for cancer have shown that it is possible to heat metal objects with pulsed electromagnetic fields selectively (PEMF), also known as induction heating. We therefore set out to answer the following research question: is non-contact induction heating of metallic implants effective in reducing bacterial load in vitro?

Objectives

Mechanical wear and corrosion at the head-stem junction of total hip arthroplasties (THAs) (trunnionosis) have been implicated in their early revision, most commonly in metal-on-metal (MOM) hips. We can isolate the role of the head-stem junction as the predominant source of metal release by investigating non-MOM hips; this can help to identify clinically significant volumes of material loss and corrosion from these surfaces.

Objectives

Third-body wear is believed to be one trigger for adverse results with metal-on-metal (MOM) bearings. Impingement and subluxation may release metal particles from MOM replacements. We therefore challenged MOM bearings with relevant debris types of cobalt–chrome alloy (CoCr), titanium alloy (Ti6Al4V) and polymethylmethacrylate bone cement (PMMA).

We implanted titanium and carbon fibre-reinforced plastic (CFRP) femoral prostheses of the same dimensions into five prosthetic femora. An abductor jig was attached and a 1 kN load applied. This was repeated with five control femora. Digital image correlation was used to give a detailed two-dimensional strain map of the medial cortex of the proximal femur. Both implants caused stress shielding around the calcar. Distally, the titanium implant showed stress shielding, whereas the CFRP prosthesis did not produce a strain pattern which was statistically different from the controls. There was a reduction in strain beyond the tip of both the implants.

This investigation indicates that use of the CFRP stem should avoid stress shielding in total hip replacement.

We dissected 20 cadaver hips in order to investigate the anatomy and excursion of the trochanteric muscles in relation to the posterior approach for total hip replacement. String models of each muscle were created and their excursion measured while the femur was moved between its anatomical position and the dislocated position. The position of the hip was determined by computer navigation.

In contrast to previous studies which showed a separate insertion of piriformis and obturator internus, our findings indicated that piriformis inserted onto the superior and anterior margins of the greater trochanter through a conjoint tendon with obturator internus, and had connections to gluteus medius posteriorly. Division of these connections allowed lateral mobilisation of gluteus medius with minimal retraction. Analysis of the excursion of these muscles revealed that positioning the thigh for preparation of the femur through this approach elongated piriformis to a maximum of 182%, obturator internus to 185% and obturator externus to 220% of their resting lengths, which are above the thresholds for rupture of these muscles.

Our findings suggested that gluteus medius may be protected from overstretching by release of its connection with the conjoint tendon. In addition, failure to detach piriformis or the obturators during a posterior approach for total hip replacement could potentially produce damage to these muscles because of over-stretching, obturator externus being the most vulnerable.

Post-mortem retrieval of canine, cemented femoral components was analysed to assess the performance of these implants in the dog as a model for human total hip replacement (THR). Mechanical testing and radiological analysis were performed to determine the stability of the implant and the quality of the cement. Thirty-eight implants from 29 dogs were retrieved after time intervals ranging from 0.67 to 11.67 years. The incidence of aseptic loosening was 63.2%, much higher than in human patients (6% in post-mortem studies). Failure of the femoral implants began with debonding at the cement-metal interface, similar to that in implants in man. The incidence of aseptic loosening was much lower in bilateral than in unilateral implants. Significant differences were observed for three different designs of implant. While the dog remains the animal model of choice for THR, results from this study provide insight into interspecies differences in the performance of implants. For example, the performance of THR in dogs should be compared with that in young rather than in elderly human patients.

Impacted morsellised allografts have been used successfully to address the problem of poor bone stock in revision surgery. However, there are concerns about the transmission of pathogens, the high cost and the shortage of supply of donor bone. Bone-graft extenders, such as tricalcium phosphate (TCP) and hydroxyapatite (HA), have been developed to minimise the use of donor bone. In a human cadaver model we have evaluated the surgical and mechanical feasibility of a TCP/HA bone-graft extender during impaction grafting revision surgery.

A TCP/HA allograft mix increased the risk of producing a fissure in the femur during the impaction procedure, but provided a higher initial mechanical stability when compared with bone graft alone. The implications of the use of this type of graft extender in impaction grafting revision surgery are discussed.

Objectives

An experimental rabbit model was used to test the null hypothesis, that there is no difference in new bone formation around uncoated titanium discs compared with coated titanium discs when implanted into the muscles of rabbits.

Dislocation remains a major concern after total hip replacement, and is often attributed to malposition of the components. The optimum position for placement of the components remains uncertain. We have attempted to identify a relatively safe zone in which movement of the hip will occur without impingement, even if one component is positioned incorrectly. A three-dimensional computer model was designed to simulate impingement and used to examine 125 combinations of positioning of the components in order to allow maximum movement without impingement. Increase in acetabular and/or femoral anteversion allowed greater internal rotation before impingement occurred, but decreases the amount of external rotation. A decrease in abduction of the acetabular components increased internal rotation while decreasing external rotation. Although some correction for malposition was allowable on the opposite side of the joint, extreme degrees could not be corrected because of bony impingement.

We introduce the concept of combined component position, in which anteversion and abduction of the acetabular component, along with femoral anteversion, are all defined as critical elements for stability.

The cortical strains on the femoral neck and proximal femur were measured before and after implantation of a resurfacing femoral component in 13 femurs from human cadavers. These were loaded into a hip simulator for single-leg stance and stair-climbing. After resurfacing, the mean tensile strain increased by 15% (95% confidence interval (CI) 6 to 24, p = 0.003) on the lateral femoral neck and the mean compressive strain increased by 11% (95% CI 5 to 17, p = 0.002) on the medial femoral neck during stimulation of single-leg stance. On the proximal femur the deformation pattern remained similar to that of the unoperated femurs.

The small increase of strains in the neck area alone would probably not be sufficient to cause fracture of the neck However, with patient-related and surgical factors these strain changes may contribute to the risk of early periprosthetic fracture.

We carried out a cross-sectional study with analysis of the demographic, clinical and laboratory characteristics of patients with metal-on-metal hip resurfacing, ceramic-on-ceramic and metal-on-polyethylene hip replacements. Our aim was to evaluate the relationship between metal-on-metal replacements, the levels of cobalt and chromium ions in whole blood and the absolute numbers of circulating lymphocytes. We recruited 164 patients (101 men and 63 women) with hip replacements, 106 with metal-on-metal hips and 58 with non-metal-on-metal hips, aged < 65 years, with a pre-operative diagnosis of osteoarthritis and no pre-existing immunological disorders.

Laboratory-defined T-cell lymphopenia was present in13 patients (15%) (CD8⁺ lymphopenia) and 11 patients (13%) (CD3⁺ lymphopenia) with unilateral metal-on-metal hips. There were significant differences in the absolute CD8⁺ lymphocyte subset counts for the metal-on-metal groups compared with each control group (p-values ranging between 0.024 and 0.046). Statistical modelling with analysis of covariance using age, gender, type of hip replacement, smoking and circulating metal ion levels, showed that circulating levels of metal ions, especially cobalt, explained the variation in absolute lymphocyte counts for almost all lymphocyte subsets.

An experimental sheep model was used for impaction allografting of 12 hemiarthroplasty femoral components placed into two equal-sized groups. In group 1, a 50:50 mixture of ApaPore hydroxyapatite bone-graft substitute and allograft was used. In group 2, ApaPore and allograft were mixed in a 90:10 ratio. Both groups were killed at six months. Ground reaction force results demonstrated no significant differences (p > 0.05) between the two groups at 8, 16 and 24 weeks post-operatively, and all animals remained active. The mean bone turnover rates were significantly greater in group 1, at 0.00206 mm/day, compared to group 2 at 0.0013 mm/day (p < 0.05). The results for the area of new bone formation demonstrated no significant differences (p > 0.05) between the two groups. No significant differences were found between the two groups in thickness of the cement mantle (p > 0.05) and percentage ApaPore-bone contact (p > 0.05).

The results of this animal study demonstrated that a mixture of ApaPore allograft in a 90:10 ratio was comparable to using a 50:50 mixture.

Impacted bone allograft is often used in revision joint replacement. Hydroxyapatite granules have been suggested as a substitute or to enhance morcellised bone allograft. We hypothesised that adding osteogenic protein-1 to a composite of bone allograft and non-resorbable hydroxyapatite granules (ProOsteon) would improve the incorporation of bone and implant fixation. We also compared the response to using ProOsteon alone against bone allograft used in isolation. We implanted two non-weight-bearing hydroxyapatite-coated implants into each proximal humerus of six dogs, with each implant surrounded by a concentric 3 mm gap. These gaps were randomly allocated to four different procedures in each dog: 1) bone allograft used on its own; 2) ProOsteon used on its own; 3) allograft and ProOsteon used together; or 4) allograft and ProOsteon with the addition of osteogenic protein-1.

After three weeks osteogenic protein-1 increased bone formation and the energy absorption of implants grafted with allograft and ProOsteon. A composite of allograft, ProOsteon and osteogenic protein-1 was comparable, but not superior to, allograft used on its own.

ProOsteon alone cannot be recommended as a substitute for allograft around non-cemented implants, but should be used to extend the volume of the graft, preferably with the addition of a growth factor.

Studies on the migration of an implant may be the only way of monitoring the early performance of metal-on-metal prostheses. The Ein Bild Roentgen Analyse - femoral component analysis (EBRA-FCA) method was adapted to measure migration of the femoral component in a metal-on-metal surface arthroplasty of the hip using standard antero-posterior radiographs. In order to determine the accuracy and precision of this method a prosthesis was implanted into cadaver bones. Eleven series of radiographs were used to perform a zero-migration study. After adjustment of the femoral component to simulate migration of 3 mm the radiographs were repeated. All were measured independently by three different observers.

The accuracy of the method was found to be ± 1.6 mm for the x-direction and ± 2 mm for the y-direction (95% percentile). The method was validated using 28 hips with a minimum follow-up of 3.5 years after arthroplasty. Seventeen were sound, but 11 had failed because of loosening of the femoral component. The normal (control) group had a different pattern of migration compared with that of the loose group. At 29.2 months, the control group showed a mean migration of 1.62 mm and 1.05 mm compared with 4.39 mm and 4.05 mm in the failed group, for the centre of the head and the tip of the stem, respectively (p = 0.001). In the failed group, the mean time to migration greater than 2 mm was earlier than the onset of clinical symptoms or radiological evidence of failure, 19.1 versus 32.2 months (p = 0.001) and 24.8 months (p = 0.012), respectively.

EBRA-FCA is a reliable and valid tool for measuring migration of the femoral component after surface arthroplasty and can be used to predict early failure of the implant. It may be of value in determining the long-term performance of surface arthroplasty.

The effects of the method of fixation and interface conditions on the biomechanics of the femoral component of the Birmingham hip resurfacing arthroplasty were examined using a highly detailed three-dimensional computer model of the hip. Stresses and strains in the proximal femur were compared for the natural femur and for the femur resurfaced with the Birmingham hip resurfacing. A comparison of cemented versus uncemented fixation showed no advantage of either with regard to bone loading. When the Birmingham hip resurfacing femoral component was fixed to bone, proximal femoral stresses and strains were non-physiological. Bone resorption was predicted in the inferomedial and superolateral bone within the Birmingham hip resurfacing shell. Resorption was limited to the superolateral region when the stem was not fixed. The increased bone strain observed adjacent to the distal stem should stimulate an increase in bone density at that location. The remodelling of bone seen during revision of failed Birmingham hip resurfacing implants appears to be consistent with the predictions of our finite element analysis.

We have investigated the role of the penetration of saline on the shear strength of the cement-stem interface for stems inserted at room temperature and those preheated to 37°C using a variety of commercial bone cements. Immersion in saline for two weeks at 37°C reduced interfacial strength by 56% to 88% after insertion at room temperature and by 28% to 49% after preheating of the stem. The reduction in porosity as a result of preheating ranged from 71% to 100%. Increased porosity correlated with a reduction in shear strength after immersion in saline (r = 0.839, p < 0.01) indicating that interfacial porosity may act as a fluid conduit.

This study explored the relationship between the initial stability of the femoral component and penetration of cement into the graft bed following impaction allografting.

Impaction allografting was carried out in human cadaveric femurs. In one group the cement was pressurised conventionally but in the other it was not pressurised. Migration and micromotion of the implant were measured under simulated walking loads. The specimens were then cross-sectioned and penetration of the cement measured.

Around the distal half of the implant we found approximately 70% and 40% of contact of the cement with the endosteum in the pressure and no-pressure groups, respectively. The distal migration/micromotion, and valgus/varus migration were significantly higher in the no-pressure group than in that subjected to pressure. These motion components correlated negatively with the mean area of cement and its contact with the endosteum.

The presence of cement at the endosteum appears to play an important role in the initial stability of the implant following impaction allografting.