The aim of this study was to systemically review the evidence for the use of intra-operative cell salvage (ICS) among patients undergoing revision hip arthroplasty, and synthesis the available data to quantify any associated reduction in allogeneic blood transfusion.

An electronic database search of MEDLINE (PubMed), EMBASE, Scopus and the Cochrane Library was completed from the date of inception to 9^th February 2021, using a search strategy and protocol created in conjunction with the PRISMA statement. Inclusion criteria were (i) adult patients >18 years, (ii) ICS utilised in one study group, (iii) revision hip arthroplasty performed. Exclusion criteria were (i) pre-donation of red blood cells, (ii) mixed reporting without dedicated subgroup analysis for revision hip arthroplasty. Screening for eligibility, and quality assessment of included studies, was performed independently by two authors (TW and DH), and any disputes settled by third author (AK).

Of the 187 records identified, 11 studies were included in the qualitative analysis, and 5 studies suitable for quantitative meta-analysis. Across the included studies there were 1856 participants, with a mean age of 63.86 years and a male: female ratio of 0.90. Quality assessment demonstrated low or medium risk of bias only. For revision hip arthroplasty patients receiving ICS, 37.07% required ABT with a mean transfusion of 1.92 units or 385ml per patient. For patients treated without ICS, 64.58% required ABT with a mean transfusion of 4.02 units or 803ml per patient. This difference achieved statistical significance (p < 0.05).

This study has demonstrated a significant reduction in the need for allogenic blood transfusion associated with intra-operative cell salvage use among patients undergoing revision hip arthroplasty. This study therefore supports the routine use of ICS in this patient group. However, a major limitation is the lack of clinical outcomes reported by the available studies. Further research is required to determine whether this effect is associated with sub-groups of revision arthroplasty procedure, and whether ICS impacts clinical outcomes such as length of stay, rehabilitation progress and mortality.

We welcome letters to the Editor concerning articles that have recently been published. Such letters will be subject to the usual stages of selection and editing; where appropriate the authors of the original article will be offered the opportunity to reply.

Aim: Since 1994, we have used virtually identical criteria to the Swedish Lumbar Spine Study Group before accepting patients for instrumented posterolateral spinal fusion. We review these criteria and the surgical outcome and compare them with those of the SLSSG. Method: 26 patients underwent posterolateral lumbar spinal fusion with Moss-Miami instrumentation between 1994 and 2001 for intractable LBP. Patients selected for surgery had to meet strict inclusion criteria. These were: severe intractable LBP causing significant work or recreational disability which failed to respond to all reasonable non-operative measures; absence of leg pain; no spondylolisthesis or spondylolysis; no neurological signs; no claudication; no other pathology i.e. mechanical/discogenic LBP only; no disc protrusion on MRI; no clinical evidence of learnt illness behaviour; L5/S1 or L4/L5 black discs on MRI, and no black discs above these levels; no litigation or compensation claim in process. Patients were assessed at a minimum of 1 year (range 14–95 months) using the NASS lumbar spine follow-up form, and postoperative radiographs to assess the fusion mass. Results: There were no deaths or significant complications of surgery. 84% of patients showed substantial improvement, 16% were improved but with significant residual pain. None were unchanged or worse. Fusion rate was 92%. Only two patients would not have undergone the same procedure again. Conclusion: We believe that these criteria form a sound basis for selecting patients with severe intractable LBP who might benefit from a posterolateral fusion.

Introduction: It is well known that the integrity of the bone cement interface is crucial for the long-term survival of a primary total hip arthroplasty (THR). Revision THR with impaction bone grafting has recently offered a solution to gross bone loss due to osteolysis. As graft becomes incorporated, clearly the bone graft/ cement interface is as crucial as the equivalent interface in primary THR. Our aim was to examine factors that inßuence this interface. Method: The study was designed to mimic clinical practice. Fresh femoral heads were harvested from primary THR. These were morcelised into large and small particles. These were characterised. The bone was impacted into a purpose built jig with measured force. Cement was pressurised onto the dried surface of the impacted bone after measured mixing times. Cement pressurisation was measured. The cement/graft specimen was extracted and transected with a band saw. Cement penetration was measured with digital image analysis. Results: Large fragment size was 29 mm², and small was 7.1mm². Light impaction was 2.2 Atm. Medium and heavy were 2.6 Atm. and 3.2 Atm. respectively. Cement penetration was inversely proportional to impaction force. Cement mixing time also signiþcantly affected cement penetration. Particle size had no effect. Conclusion: Allograft should be adequately but not excessively impacted, to allow good cement incursion. Cement should be introduced and pressurised perhaps as early as two minutes. Fragment size does not affect cement penetration.

Aim: A ten-year review of uncemented hip arthroplasty using the Spotorno (CLS) stem, Morscher (Mathys) cup and a ceramic head in patients under the age of 60. Method: 113 hips implanted in 104 patients (mean age 50.5: range 19–60) between 1987 and 1995. Mean follow-up: 9.7 years. A postal questionnaire, based on the modiþed Harris Hip Score was followed by clinical and radiographic examination. Results: 3 patients had died and 18 (17%) could not be traced. Complications: sepsis 2%; aseptic loosening 5%; gross polyethylene wear 1%; fractured ceramic head 1%; periprosthetic fracture 1%. Pain was ñabsentñ or ñmildñ in 90%,and ñmoderateñ in 10%. 59.7% had a modiþed Harris Hip Score above 90, 35.8% between 70 and 90, and 4.5% less than 70 (mean 89.4%). According to Enghñs criteria, femoral stems had Ôingrowth conþrmedñ in 94%, suspected in 5% and þbrous encapsulation in 1%. None was predicted as unstable. Osteolysis was absent around 65.8% of stems, conþned to the neck in 14.8%, present proximally in 17.1% and distally in 2.3%. According to ARA criteria, acetabular þxation was Ôexcellentñ in 72.1%, Ôfairñ in 12.8% and Ôpoorñ in Osteolysis around the cup was Ôabsentñ or in Ôminimal in 91.8%, Ômoderateñ in 3.5% and Ôsevereñ in 4.7% Conclusions: We observed an association between acetabular polyethylene wear rates and radiographic evidence of osteolysis, which was greater around the rough-blasted stem than the hydroxyapatite-coated cup. Where osteolysis is present, the linear wear rate is consistently greater than 0.1mm/year, which may seem high but is comparable to other currently-used cement-less combinations.

Aims: To evaluate the radiological evidence of graft incorporation in the femur following impaction bone grafting. Methods: Consecutive patients that underwent femoral impaction bone grafting using two stem designs were identified. Radiographs were taken post-op and at 3, 6 and 12 months then yearly. They were evaluated for radiological signs of graft incorporation and the results plotted against time and according to Gruen zones. Results: 28 patients over a 4-year period underwent femoral impaction grafting. 16 had a double tapered polished stem (Exeter), 12 had a non-polished stem (Charnley Elite). There were 17 male and 11 female patients with an average age of 62.6 years.

There is little radiological evidence of incorporation before 6 months, but by one year 75% of patients show some graft incorporation in some or all Gruen zones. Graft incorporation in the Exeter stems is seen first in zones 4 and 5. In the Charnley stems it is first seen in zones3, 5 and 6. The majority of subsidence occurs within the first year following surgery. Conclusions: Graft incorporation is seen first in those zones that may be expected to experience greater load. Differences in the pattern of graft incorporation may reflect differences in stress distribution between the two stem designs. Some load appears to promote graft incorporation. The construct must have sufficient stability to withstand subsidence whilst graft incorporation occurs.

Between 1969 and 1985 26 patients with destructive lesions of the distal humerus were treated by endoprosthetic replacement; each implant was custom-made and incorporated part of the distal humerus or the entire bone as well as a hinged total elbow replacement. Recurrence occurred in three of the patients with tumours, and three prostheses were removed because of deep infection in patients with previously compound injuries of the elbow. Another three loosened without infection, but none needed revision or removal and no amputations resulted. Other complications included nerve palsies, but the only deaths were from metastases. A useful range of elbow movement, with a stable arm and good hand function, was achieved in every patient.

Endoprosthetic replacement of the proximal humerus has been performed in our unit on 25 occasions between 1950 and 1982. The indication for surgery was destruction of the proximal half of the humerus so extensive that the only alternatives were reconstruction or amputation. Of the patients with tumours two died from metastases, and three from unrelated causes; local recurrence necessitated amputation in two patients. Minor complications were frequent, but there were no deep infections and, after 1964, no prosthesis became loose. Active shoulder movement after operation was considerably limited, but passive movement was good and function of the elbow and hand were preserved.

Structural scoliosis occurs more commonly in patients with juvenile chronic arthritis than in the normal population. We have reviewed 32 patients with both juvenile arthritis and a scoliosis and suggest that structural curves may arise from postural curves associated with asymmetrical involvement of lower limb joints.