Introduction

Modern processing techniques in bone banking are thought to decrease the presence of allogenic material in bone. This project was performed to observe any changes in peripheral blood lymphocyte subsets in response to allografted bone used in revision hip replacement.

Periprosthetic femoral fractures around total knee arthroplasty present a challenge in octogenarians with advanced osteoporosis. We describe a salvage technique combining retrograde intramedullary nailing augmented with polymethylmethacrylate (PMMA) cement in five patients followed up for a median time of 12 months. The nail/cement construct bridges the femoral canal tightly and behaves like a stemmed cemented revision component. All patients had an uncomplicated recovery and returned to their pre-injury functional status within four months. This procedure does not disrupt the soft tissue envelope facilitating periosteal callus formation, is easy to perform and permits immediate full range of movement. When standard retrograde nailing or plating alone is inadequate in maintaining severely osteoporotic fracture reduction in octogenarians unfit for lengthy procedures, nailed cementoplasty is proposed as a salvage procedure.

This paper could not be resubmitted: The paper was initially presented at the BHS and they have already published it in the JBJS Supplement.

Introduction: Freezing and storage of fresh frozen femoral heads destined for use in revision hip arthroplasty is thought to result in graft cell death. Washing of the graft following the morsellisation process also removes a large proportion of the marrow content of the allograft. However, the immunological load of the impaction allografting process remains unknown. The aim of this study was to investigate the immune response by observing any changes in peripheral blood lymphocyte subsets in response to allografted bone used in revision hip replacement

Methods: 87 patients were entered into this prospective study and grouped according to whether impaction allograft was used or not. Venous blood samples were collected pre-operatively and at set time intervals up to one year post-operatively. Using flow cytometry, analysis of venous blood allowed counts of the following cells: Helper T-lymphocytes, cytotoxic T-lymphocytes, memory T-lymphocytes, naïve T-lymphocytes, Natural Killer cells and B-lymphocytes.

Results: All patients had a successful outcome at one year. 50 patients with radiologically defined host-graft union were compared with 37 patients who did not receive allograft. Pre-operatively, a significant difference (p=0.03) was found between the groups of patients with respect to Natural Killer cells but other subsets showed no significant difference. Post-operatively the significant difference between Natural Killer cells resolved. T-helper lymphocytes, cytotoxic lymphocytes, memory T-lymphocytes and naïve T-lymphocytes in both groups showed decreases in values immediately post surgery, recovering to normal values within 6 weeks post-surgery. The allograft group showed significant increases from baseline in cytotoxic T-lymphocytes at 6 months (p< 0.01) and memory T-lymphocytes one year postoperatively (p=0.04). B-lymphocyte numbers did not alter significantly from baseline.

Conclusion: Cytotoxic T-lymphocytes recognise HLA-class I molecules which are present on all nucleated cells and have been implicated in having a role in osteoclast regulation. Memory T-lymphocytes are produced after a naïve T-lymphocyte is exposed to an antigen. The observed increases of these subsets were not observed in the non-grafting group suggesting the allografted bone had elicited an immunological response. At 12 months all grafts appeared radiologically stable and the immunological response may have been beneficial to the outcome.

Introduction: The long term success of impaction grafting depends on the remodelling process during incorporation. This project was designed to characterise any differences in the biochemical markers of bone turnover following revision hip arthroplasty performed with or without impaction grafting.

Methods: 87 patients undergoing revision hip arthroplasty were entered into this prospective study and grouped according to whether impaction allograft was used or not. Biochemical markers of bone turnover were assessed pre-operatively and post-operatively on day 2, day 9, week 6, 6 months and 1 year. Osteocalcin, pro-collagen type-I N-terminal propeptide and bone specific alkaline phosphatase were measured as bone formation markers. C-telopeptide, pyridinoline and deoxypyridinoline were measured as bone resorption markers.

Results: All patients had a successful radiological and clinical outcome at one year. 50 patients with radiologically defined host-graft union were compared with 37 patients who did not receive allograft. Markers of bone formation tended to rise by day 9 but the rise in osteocalcin was delayed in the graft group and was significantly lower at 6 months in comparison to the non-graft group (p=0.002). Alkaline phosphatase levels remained significantly elevated at one year in the graft group (p=0.027) whilst levels in the non-graft group had normalised. Markers of bone resorption also rise in both groups but with no significant differences between the groups.

Discussion: Following impaction grafting, new bone formation may be delayed in comparison to revisions performed without graft. The pattern of markers of bone resorption did not differ significantly between the groups suggesting that there is no large scale resorption of the impacted allograft in these cases.

These results provide a biochemical insight into the bone formation and bone resorption processes during allograft incorporation.

Introduction: Septic arthritis of the hip is an Orthopaedic emergency. While common in the paediatric population, it is rare in adults and little is reported on it.

Methods: Retrospective review of cases presenting to Leeds General Infirmary, St James’s University Hospital & York District Hospital over a fourteen year period from 1991–2005.

Results: 46 cases identified. The mean age was 43 years and 80% (37/46) of cases were male. Risk factors for sepsis were present in 87% (40/46) and consisted of IVDU in 48% (22/46), DM in 20% (9/46), liver disease in 22% (10/46) and immunosuppressive drugs in 17% (8/46). Rheumatic joint disease was present in 28% (13/46) consisting of rheumatoid arthritis, gout and psoriatic arthropathy in 13% (6/46), 9% (4/46) and 7% (3/46) respectively. The primary sources of sepsis were IVDU in 48% (22/46), unknown in 39% (18/46), line sepsis in 11% (5/46) and psoas abscess in 2% (1/46). All cases presented with the triad of groin pain, constitutional upset and difficulty or inability to weight bear. Examination revealed pyrexia in 57% (26/46) and a painfully decreased range of joint motion in all cases. Laboratory tests revealed a raised white cell count in 57% (34/46) and raised CRP’s and ESR’s in all cases. The most commonly isolated organism was staphylococcus aureus 78% (36/46). Medical treatment consisted of antimicrobial therapy for a mean length of 49 days. Surgical treatment consisted of a mean of 1.8 arthrotomies or arthroscopic hip washouts and debridements in consenting cases- 89% (41/46). Two stage total hip arthroplasty for symptom control was required in 4% (2/46). Outcomes were good with sepsis related mortality of 4% (2/46) and local recurrence of 9% (4/44).

Discussion: Septic arthritis of the hip is a potentially serious condition. Timely medical and surgical intervention can lead to good outcomes.

Introduction: The long term success of impaction grafting depends on the remodelling process during incorporation. This project was devised to characterise any differences in the biochemical markers of bone turnover following revision hip arthroplasty performed with or without impaction grafting.

Methods: 87 patients were entered into this prospective study and grouped according to whether impaction allograft was used or not. Biochemical markers of bone turnover were assessed pre-operatively and post-operatively on day 2, day 9, week 6, 6 months and 1 year. Osteocalcin, procollagen type-I N-terminal propeptide and bone specific alkaline phosphatase were measured as bone formation markers. C-telopeptide, pyridinoline and deoxypyridinoline were measured as bone resorption markers.

Results: All patients had a successful outcome at one year. 50 patients with radiologically defined host-graft union were compared with 37 patients who did not receive an allograft. Markers of bone formation tended to rise by day 9 but the rise in osteocalcin was delayed in the graft group and was significantly lower at 6 months in comparison to the non-graft group (p=0.002). Alkaline phosphatase levels remained significantly elevated at one year in the graft group (p=0.027) whilst levels in the non-graft group had normalised. Markers of bone resorption also rise in both groups but with no significant differences between the groups.

Discussion: Following impaction grafting, new bone formation may be delayed in comparison to revisions performed without graft. The pattern of markers of bone resorption did not differ significantly between the groups suggesting that there is no large scale resorption of the impacted allograft in these cases.

These results provide a biochemical insight into the bone formation and bone resorption processes during allograft incorporation.

Factors influencing the results of revised cemented sockets with bone grafting have been studied in 249 cases.

Freeze-dried allografts in 77 and fresh frozen in 172 cases have been used. The average follow-up was 8 years 11 months for the freeze-dried group and 2 years 11 months for the fresh frozen cases. There were 13 postoperative dislocations, 20 TNU, 4 thromboembolic complications, 4 delayed wound healing and 2 intraoperative fractures of the acetabulum. There have been 11 re-revisions: 8 for aseptic loosening, 2 for dislocation and 1 for infection. Radiographic evidence of loosening was seen in another 38 cases.

The acetabular bone stock at the time of revision and initial stability of socket fixation had a significant influence on the outcome. Direction of socket migration before surgery appeared to predict risk of failure. The primary pathology, type of bone graft and grafting technique also had an effect.

We previously reported the result of 45 Charnley LFA’s with femoral head autograft for Developmental Dysplasia of the hip with a minimum follow-up of ten years.

After an average follow-up of eleven years there was no revision. One socket migrated and four sockets were fully demarcated.

To assess our long-term results we reviewed the clinical and radiological findings in the same group of patients that had been studied previously.

To date 5 patients died from causes unrelated to the hip replacement and were excluded from the final radiological analysis. 40 Charnley LFA’s have been followed-up regularly.

The average follow-up is now 17 years 1 month / range: 15–21 years/. Three sockets have been revised: two for aseptic loosening and one for infection. Radiographic assessment showed that three sockets migrated and four had full demarcation.

Demarcation at the cement-bone interface of the socket was rare in zone one but was common in zone two.

We concluded that sound fixation of the autograft and orientations of the acetabular component are essential. We recommend that solid bone graft should be combined with impaction bone grafting in dysplastic cases. We also observed that bone grafting at primary surgery gives better chances for component fixation at the time of revision.

We report the result of 49 revisions for aseptic cup loosening using freeze-dried allografts.

We assessed the results according to the primary pathology, severity of bone loss, direction of socket migration before revision, method of bone grafting, socket position, graft incorporation and socket loosening.

Patient’s mean age at revision was 56 years 8 months. At a mean follow up of 7 years and 5 months four cups have been revised for aseptic loosening. Five sockets have migrated but remained stable while six showed full demarcation but have not migrated.

The other 34 remained stable and show radiographic evidence of graft incorporation.

The calcar femorale is a vertical plate of bone lying deep to the lesser trochanter and is formed as a result of traction of the iliopsoas which separates the femoral cortex into two distinct layers, the calcar femorale and the medial femoral cortex. They fuse together proximally to form the medial femoral neck. A stem placed centrally will abut against the calcar femorale with little or no space for cement. Clearing of the calcar will offer space for a cement layer, which will support the stem proximally on the posterior aspect. We compared two consecutive groups of Charnley low-friction arthroplasties, with and without clearing of the calcar.

In 330 patients who had an arthroplasty without clearing the calcar, there were ten revisions for aseptic loosening of the stem and six other stems were considered ‘definitely loose’, giving a rate of failure of 4.8%. In 111 patients in whom the calcar was cleared there was only one revision for aseptic loosening and no stems were classed as ‘definitely loose’, giving a rate of failure of 0.9%.

Survivorship analysis has again shown the need for long-term follow-up; the differences became clear after ten years but because of the relatively small numbers, statistical analysis is not yet applicable.

We now clear the calcar femorale routinely and advocate optimal access to the medullary canal and insertion of the stem in the area of the piriform fossa.

Between 1983 and 1988 we carried out 45 Charnley low-friction arthroplasties with autografts from the femoral head in 41 patients for developmental dysplasia of the hip.

The preoperative radiographs were assessed for the severity of DDH according to the classifications of Crowe et al, Hartofilakidis et al and Sharp. The postoperative and follow-up radiographs were analysed for coverage of the socket by the graft, for loosening and for the outcome of the fixation of the bone graft. Two patients died (two hips) at four and seven years after THR from causes unrelated to the surgery and were excluded from the final radiological analysis. The mean age of the patients at the time of operation was 46 years 3 months. The autograft of the femoral head covered a mean 26% (16 to 35) of the acetabular component. All the grafts united. Some degree of resorption of the bone graft occurred in 27 patients, and always involved the lateral part of the graft, which was beyond the margin of the socket. After a mean follow-up of 11 years there had been no revisions and 38 patients had no pain or only slight discomfort. One socket migrated and four others were fully demarcated.

Our findings indicate that the Charnley LFA with an autograft of the femoral head for DDH remains successful at a follow-up of 15 years.