Introduction: Debate continues regarding the optimal timing of surgery for patients requiring bilateral knee arthroplasty; we reviewed the costs, clinical and functional outcomes of 116 patients undergoing simultaneous or staged bilateral surgery using 3 different prostheses.

Method: Data were retrospectively collected from 116 consecutive patients undergoing 232 knee replacements over 10 years, either simultaneously or over 2 hospital admissions (staged). Post-operative complication rate, total cost of treatment (calculated from pros-theses, theatre time, days in hospital and number of clinic attendances) and functional (HSS) score at 1 year were the outcome measures.

Results: 54 patients underwent Oxford unicondylar knee replacements, 41 simultaneously, 13 staged; respective mean total costs were £9890 and £13,553 (p< 0.001). 42 patients were treated with AGC prostheses; 14 simultaneously, 28 staged, with respective total costs of £12,187 and £16,920 (p< 0.001). 10 TMK patients had simultaneous surgery (mean total cost £14,812), 10 were staged (£20,191); p< 0.001. For all 3 prostheses, there was no significant difference in complication rate or 1 year functional outcome between simultaneous and staged groups.

Discussion: Some authors advocate replacing both knees simultaneously as safe and cost-effective; others report significant increases in medical and surgical morbidity. Our series shows significant cost savings with no increase in complication rate.

Aim: To establish the effectiveness of using the intra-operative cell saver in spinal surgery.

Methods: Patients undergoing posterior instrumental lumber spine fusion with iliac crest bone graft were selected to have intra-operative red cell salvage using the cell saver machine (Dideco Electra-Auto-transfusion Cell Separator). 20 patients were in the study group. The control group consisted of 28 patients who had undergone similar surgery prior to introducing the cell saver. The parameters identified were: pre and post operative haemoglobin, clotting state, volume of transfused allogenic blood, volume of transfused autollogous blood using the cell saver and indications for transfusion. Statistical analysis: the chi-square and the t-test.

Results: The average age in the cell saver group was 43.8 years and in the control group 48.3 (p> 0.09). The number of levels fused was comparable between the two groups (p> 0.1). There was no difference in the pre and post operative haemoglobin level in the two groups (p> 0.7 & p> 0.3 respectively). No patient had a pre-operative coagulopathy. Two patients (10%) in the cell saver group received an intra-operative allogenic transfusion, 14 patients (50%) in the non-cell saver group received a transfusion. The difference was significant (p< 0.004). Conclusion: The use of the cell saver significantly reduces the need for allogenic blood transfusion in major spinal surgery. We therefore recommend its routine use in such procedures.

Aim of the study: To determine the optimum hydroxy-apatite-coating pattern of the femoral component in cementless hip arthroplasty.

Methods: Between January 1996 and May 1997, the first 65 patients (73 hips) of a single center prospective trial were randomised to receive a proximally coated Osteonics or a fully coated JRI Furlong stem. 30 Osteonics and 43 JRI stems were implanted. Clinical assessment was carried out using the modified Merle D’Aubigne and Postel (MDP) scoring system and the visual analogue scale (VAS). Radiological evaluation included measurement of subsidence, bone resorbtion, pedestal formation and implant-bone interface assessment. All live patients were reviewed. The average follow up was 50.3 months in the JRI group and 51.8 months in the Osteonics group.

Results: There were eight intra-operative femoral and one medial acetabullar wall fractures in the JRI group. In the Osteonics group there was one intra operative femoral fracture. There was one cup revision in both groups for aseptic loosening and one excision arthroplasty in the JRI group for sepsis. Clinical evaluation revealed a mean MDP of 16.3 (8–18) and a mean VAS of 0.55 (0–7) for the JRI group. The mean MDP for the Osteonics group was 16.31 (8–18) and the mean VAS 0.62 (0–4). Radiological evaluation revealed resorbtion in zone 7A in 16 of the JRI hips and 5 of the Osteonics. There was a reactive line around the distal part of the Osteonics stem in 17 patients, which was not correlated with symptoms. Subsidence rates were limited and comparable.

Conclusion: Both hips performed well. There was no difference in revision rate for loosening (p> 0.84). Clinical evaluation showed no significant difference in outcome between the prostheses (p> 0.83 for MDP & VAS). Radiology revealed different patterns, which did not represent a clinical importance at this stage.

Aim of the study: To prove that tapered pegs are effective in reducing tibial tray subsidence in vitro and that this effect is related to the dimensions of these pegs.

Methods: The peg designs were drawn up mathematically to allow for a unified surface area – Three different designs were used. The pegs were made from cobalt chrome, were conical in shape with a cut off tip and had a variable base and height and an equal surface area. These pegs were fixed with screws to an IB 11 HA coated tibial tray.

Wet foam was used as bone substitute, this is an open cell foam that is fairly fragile but has the benefit of being constant and is cheap and readily available. This foam is not desired to have cancellous bone characteristics but is useful in observing the relative effect of adding these pegs. Two different settings in vivo were mimicked: that of a tibial tray and pegs resting fully on cancellous bone, in which case a central vertical force was applied, and that of the tray resting on the cortex on one side with a lateral vertical force applied over the other side in both the proud and flush setting (2& 4 pegs respectively). The investigation was undertaken using a home made system allowing a crude estimate of the forces producing initial subsidence, which was identified by initial fracture of the foam, and total subsidence which was identified as total failure of the foam. Each test was carried out three times. Controls were carried out on the tray with no pegs and on the pegs individually before attaching these to the tray and repeating the tests for each design.

Results: Using this crude approach, the mean control force for total subsidence of the pegs was as follows: Short with wide base 550.3 g (± 45.3 g), medium length and base 475.6 g (± 24.25 g), long with narrow base 364.5 g (± 24.25 g). The mean control force for initial subsidence of the tray without pegs when subjected to a vertical central force was 4.3 kg (4–4.5 ± 0.27 kg) and the total subsidence force for the tray was 7.32 Kg (5.5–8, ± 0.84 kg). The mean central vertical force for initial subsidence of the tray with the tapers mounted was 7.16 kg (7–7.5 ± 0.28), for the short wide pegs, 5.33 kg (5–5.5 ± 0.28) for the medium pegs and 5.33 kg (5–6 ± 0.57) for the long pegs. The mean central vertical force for total subsidence of the tray with the tapers mounted was 9 kg (8.5–9.5 ± 0.5) for the short wide pegs, 9.8 kg (8–11 ± 1.6) for the medium pegs and 9.6 kg (8.5–11.5 ± 1.6) for the long pegs. The mean lateral control force for total subsidence of the proud tray with pegs resting on the wooden ledge was 5 kg (4–6 ± 0.75). The mean lateral vertical force for total subsidence with all pegs mounted was 7.16 kg (7–7.5 ± 0.28) for the short pegs, 5.8 kg (5.5–6 ± 0.28) for the medium pegs and 5.5 (5.5–5.6 ± 0.05) for the long pegs. No definite initial subsidence force could be identified. The mean lateral control force for total subsidence of the flush tray resting on the wooden ledge was 13.16 kg (12.5–14 ± 0.76).

The mean lateral vertical force for total subsidence with pegs mounted on the foam side was 12.3 kg (11.5–13 ± 0–76) for the short pegs, 13.5 kg (12–15.5 ± 1.8) for the medium pegs and 13.83 kg (12–15.5 ± 1.7) for the long pegs. Again no definite initial subsidence force could be identified.

Conclusion: The addition of tapered conical pegs to the tibial tray increases the resistance to subsidence when subjected to a central vertical force with the tray sitting fully on foam. The initial subsidence resistance was more marked in the case of the short wide variety. In the case of the tray resting on the hard edge and a lateral force applied, the proud tray showed improved resistance to total subsidence with the short pegs while the flush tray did not show improvement with pegs and was marginally worse with the short pegs. This is probably due to a higher margin of observer error.