Search

The disadvantages of allografts to restore femoral bone defects during revision hip surgery have led to the search for alternative materials. We investigated the feasibility of using porous titanium particles and posed the following questions: (1) Is it possible to create a high-quality femoral graft of porous titanium particles in terms of graft thickness, cement thickness, and cement penetration? (2) Does this titanium particle graft layer provide initial stability when a femoral cemented stem is implanted in it? (3) What sizes of particles are released from the porous titanium particles during impaction and subsequent cyclic loading of the reconstruction? We simulated cemented revision reconstructions with titanium particles in seven composite femurs loaded for 300,000 cycles and measured stem subsidence. Particle release from the titanium particle grafts was analyzed during impaction and loading. Impacted titanium particles formed a highly interlocked graft layer. We observed limited cement penetration into the titanium particle graft. A total mean subsidence of 1.04 mm was observed after 300,000 cycles. Most particles released during impaction were in the phagocytable range (< 10 microm). There was no detectable particle release during loading. Based on the data, we believe titanium particles are a promising alternative for allografts. However, animal testing is warranted to investigate the biologic effect of small-particle release.

Bone impaction grafting has been advocated for reconstruction of femoral head osteonecrosis. However, bone grafts do not prevent the progression of collapse and subsequent disabling osteoarthritis in late-stage osteonecrosis. We hypothesized reconstruction of large subchondral defects with a mix of impacted morsellized cancellous bone grafts and calcium phosphate cement would afford mechanical stability allowing remodeling into new bone without femoral head collapse. In a pilot study we created a critically sized subchondral defect model in 15 goats following the trapdoor procedure. Defects were left empty (n = 3), filled with morsellized cancellous bone grafts (n = 6), or filled with a mixture of morsellized cancellous bone and calcium phosphate cement (n = 6). No defects collapsed by sacrifice at 12 weeks. Defects filled with morsellized cancellous bone showed complete incorporation and remodeling to a normal trabecular structure. In the morsellized cancellous bone/calcium phosphate cement group, most of the calcium phosphate cement was resorbed in 12 weeks and the mixture was largely replaced by fibrous or fatty marrow. Although the specific mixture did not confirm our hypothesis, we suspect a slower resorbing calcium phosphate cement remains a promising material to mix with morsellized cancellous bone to treat late stages of femoral head osteonecrosis.

Segmental defects can be reconstructed with a cortical strut or a metal wire mesh when using bone impaction grafting in the femur. We hypothesized that structural grafts would negatively influence revascularization of the underlying impacted grafts compared with an open wire mesh. A standardized large medial wall defect was reconstructed with a strut or a mesh in six goats per group. In all femurs impaction grafting was done in combination with a cemented collarless double-tapered highly polished Exeter stem. After 6 weeks the femurs were harvested. A high rate of periprosthetic fractures was observed (three of seven and two of six for the strut and mesh groups, respectively). Histologic analysis showed different revascularization and tissue ingrowth patterns for both reconstruction techniques. In the strut group, fibrous tissue ingrowth was limited to the edges of the defect. Medially behind the strut no or limited fibrous tissue ingrowth was found. In the mesh group, fibrous tissue and blood vessels penetrated the mesh and a superficial zone of revascularized grafts was observed. Although revascularization, concomitant graft resorption and bone incorporation may compromise the short-term stability of the stem after surgery, the long-term stability of the stem probably is best guaranteed by graft incorporation.

In a one-surgeon study the clinical and radiographic results of 30 cementless bipolar hip prostheses in 24 patients younger than 55 years were evaluated. Eleven noncoated prostheses (Noncoated Group) and 19 hydroxyapatite-coated prostheses (Hydroxyapatite Group) were compared after a mean followup of 10.4 years. The Harris hip score increased from a preoperative average of 41 points to 76 points at final followup (Noncoated Group, 70 points; Hydroxyapatite Group, 80 points). Thigh or groin pain was present in 15 patients (16 hips) (Noncoated Group, 55%; Hydroxyapatite Group, 53%). Radiographically, subsidence was the major problem at the noncoated prostheses (Noncoated Group, 91%; Hydroxyapatite Group, 5%), whereas osteolysis mainly was seen at the hydroxyapatite-coated prostheses (Noncoated Group, 18%; Hydroxyapatite Group, 89%). The obtained aseptic revision rate of 27% (Noncoated Group, 27%; Hydroxyapatite Group, 26%) is too high to use this implant in young patients. The large amounts of polyethylene wear debris generated by the bipolar system play an important role in this limited success. In the initially well-fixed hydroxyapatite-coated prostheses the sealing effect of a hydroxyapatite coating creates high concentrations of polyethylene in the limited joint space, resulting in massive proximal femoral osteolysis. Consequently, a hydroxyapatite coating introduces a new failure mechanism. Therefore, hydroxyapatite does not improve the outcome of a cementless bipolar hemiarthroplasty in the long-term.

Even in extensive osteonecrosis of the femoral head in younger patients, a femoral head-preserving method is preferable. We developed a new technique using the lateral approach as used in traditional core biopsy; the osteonecrotic lesion was removed and impacted bone grafts were used to regain sphericity and prevent collapse. In this prospective one surgeon study, we included 28 consecutive hips in 27 patients with extensive osteonecrotic lesions (ARCO classification Stage 2 [11 hips], Stage 3 [14 hips]; and Stage 4 [three hips]); 14 hips had preoperative collapse. The mean age of the patients was 33 years (range, 15-55 years). At a mean followup of 42 months (range, 24-119 months), eight hips (29%) were converted to a total hip arthroplasty (THA). Of the 20 reconstructions that were in situ, 18 were clinically successful (90%) and 70% were radiologically successful. Patients who were younger than 30 years at surgery had a radiologically significant better outcome, even patients with higher stages of osteonecrosis. Patients with preoperative collapse and use of corticosteroids had disappointing results. This method is attractive as a salvage procedure, is relatively simple and quick, and it does not interfere with an eventual future hip arthroplasty.

Bone defects in revision hip surgery can be reconstructed with impacted morselized bone grafts. Rinsing these trabecular allografts may enhance graft incorporation by washing out immunogenic factors present in blood, marrow, and fat. However, it has been proposed that impaction of the graft releases biologically active factors, which can provide sufficient activity to stimulate new bone formation. Rinsing before impaction could enhance bone allograft incorporation, but rinsing after impaction could diminish the incorporation process of impacted bone graft. To study the effect of rinsing and impaction of morselized bone grafts on bone ingrowth, a bone chamber study was done in goats. Autografts and allografts were divided into three treatment groups: (A) impacted; (B) rinsed and impacted; and (C) rinsed, impacted, rinsed, and impacted again. Ten goats received three bone chambers in each proximal tibia. The chambers were filled with either allograft or autograft, yielding six different implants per goat. After 6 weeks, histologic analyses were done and bone and tissue ingrowth were measured. New bone and total tissue ingrowth were higher in autografts than in allografts, especially in the nonrinsed group. With rinsing, total tissue ingrowth increased in the allograft group to approach that of autografts. Rinsing after impaction did not additionally alter bone ingrowth. The current findings show that incorporation of allografts can be improved by rinsing the grafts before impaction.

Animal studies have shown almost complete incorporation of impacted morselized bone grafts. To determine whether this also is true in humans, 24 acetabular bone biopsy specimens from 21 hips of 20 patients were examined. Biopsy specimens were obtained 3 months to 15 years after acetabular reconstruction in primary and revision total hip arthroplasties in combination with a cemented cup. Histologic examination showed rapid revascularization of the graft, directly followed by osteoclastic resorption and woven bone formation on the graft remnants. New bone also was formed on fibrin accumulations or without any scaffold in the fibrous stroma tissue that had invaded the graft. Thereafter the mixture of graft, new bone, and fibrin was remodeled completely into a new trabecular structure, with normal lamellar bone and only scarce remnants of graft material. Localized areas of nonincorporated bone graft surrounded by fibrous tissue remained, irrespective of the followup period. Large nonincorporated fragments of cartilage also were found, particularly in cases in which femoral head bone chips were produced by a bone mill. In general, impacted trabecular bone chips incorporate by a mechanism that is similar to that previously observed in animal studies. In some patients, however, areas of nonincorporated bone graft remained and long-term alterations were found, probably related to the loosening process.

Acetabular bone stock loss compromises the outcome in primary and revision total hip arthroplasty. In 1979, a biologic method was introduced with tightly impacted cancellous allografts in combination with a cemented polyethylene cup for acetabular reconstruction. With this technique, it is possible to replace the loss of bone and to repair normal hip mechanics and hip function with a standard implant. Based on the authors' 20 years experience, a review of the long-term data is presented in primary total hip arthroplasty with preexisting acetabular bone stock loss, primary total hip arthroplasty in rheumatoid arthritis, patients who had bone impaction when younger than 50 years, and in acetabular revisions. The survival rate with revision of the cup for aseptic loosening as the end point was 94% at 10 to 17 years, 90% at 10 to 18 years, 91% at 10 to 17 years, and 92% at 10 to 15 years. From biopsy specimens from humans and histologic data in animal experiments the incorporation of these impacted bone chips was proven. The acetabular bone impaction technique using large morselized bone chips (range, 0.7-1 cm) and a cemented cup is a reliable technique with favorable long-term outcome.

Revision of a failed total joint replacement often demands bone grafting methods to restore deficient bone stock. However, impaired allograft incorporation can be the result of inadequate host or graft properties. The stimulation of bone healing with growth factors could provide a new approach to deal with this problem. The repeated sampling bone chamber was used in the goat to investigate the properties of bone allografts enriched with transforming growth factor-beta, recombinant human bone morphogenetic protein-2, and basic fibroblastic growth factor under unfavorable vascular and nonloaded conditions. Ten goats each had three bone chambers implanted in the medial proximal tibia. Different carrier allograft bone preparations were used for each growth factor based on convention and previously reported results. The period between implantation and chamber harvest was 8 weeks. The concentrations of the growth factors used was 0, 1, or 10 micrograms of transforming growth factor-beta 2, 0, 1, or 5 micrograms of bone morphogenetic protein-2, and 0, 40, 200 ng of basic fibroblastic growth factor. The specimens were analyzed histomorphometrically for the amount of soft tissue ingrowth, bone ingrowth, and the number of osteoclasts. In all specimens, a resorption front grew into the graft followed by fibrovascular tissue and, in some instances, bone. In the 5 micrograms of bone morphogenetic protein-2 specimens, larger amounts of soft tissue and woven bone were present, whereas in the specimens that received 10 micrograms of transforming growth factor-beta 2, there was a decrease in the amount of tissue and bone ingrowth. Two hundred nanograms of basic fibroblastic growth factor had a negative effect on soft tissue formation but increased the amount of vascular elements containing erythrocytes. The number of osteoclasts was higher in the 5-microgram bone recombinant human morphogenetic protein-2 specimens. In the clinical arena with absence of good perigraft vascularization and loading, bone morphogenetic protein-2 may have a strong stimulatory effect on bone graft incorporation.

Loosening of primary cemented and noncemented components of total hip arthroplasties always is accompanied by a loss of bone stock. There are several options for reconstruction of the acetabular and femoral defects. The authors' treatment of choice is a standardized cemented revision procedure with tight impaction of morsellized cancellous autograft or allograft chips in acetabular and femoral reconstructions. In this study, the clinical and radiographic evaluation of acetabular defects reconstructed with impacted morsellized allograft femoral heads was described. A cemented cup supplemented with morsellized cancellous grafts and wire meshes in cases of segmental defects was sufficiently stable to allow for complete graft consolidation. After a mean followup of 70 months of 88 hips, 4 cases of clinical failures (including 1 infection) and 6 cases of radiologic failure of the reconstructions were observed, resulting in a failure percentage of 11.4% after 5 years. Autografts and allografts were equally effective. Because the clinical success of the technique also was supported by the results of histologic and biomechanic studies in animals, the authors were encouraged to continue this technique, not only in the acetabulum, but also in the femur.