The management of bone loss in revision replacement of the knee remains a challenge despite an array of options available to the surgeon. Bone loss may occur as a result of the original disease, the design of the prosthesis, the mechanism of failure or technical error at initial surgery. The aim of revision surgery is to relieve pain and improve function while addressing the mechanism of failure in order to reconstruct a stable platform with transfer of load to the host bone. Methods of reconstruction include the use of cement, modular metal augmentation of prostheses, custom-made, tumour-type or hinged implants and bone grafting.

The published results of the surgical techniques are summarised and a guide for the management of bone defects in revision surgery of the knee is presented.

Between 1976 and 2004, 38 revision arthroplasties (35 patients) were performed for aseptic loosening of the humeral component. The mean interval from primary arthroplasty to revision was 7.1 years (0.4 to 16.6). A total of 35 shoulders (32 patients) were available for review at a mean follow-up of seven years (2 to 19.3).

Pre-operatively, 34 patients (97%) had moderate or severe pain; at final follow-up, 29 (83%) had no or only mild pain (p < 0.0001). The mean active abduction improved from 88° to 107° (p < 0.01); and the mean external rotation from 37° to 46° (p = 0.27). Excellent or satisfactory results were achieved in 25 patients (71%) according to the modified Neer rating system. Humeral components were cemented in 29, with ingrowth implants used in nine cases. There were 19 of standard length and 17 were longer (two were custom replacements and are not included). Bone grafting was required for defects in 11 humeri. Only two glenoid components were left unrevised. Intra-operative complications included cement extrusion in eight cases, fracture of the shaft of the humerus is two and of the tuberosity in four. There were four re-operations, one for recurrent humeral loosening, with 89% survival free of re-operations at ten years.

Revision surgery for aseptic loosening of the humeral component provides reliable pain relief and modest improvement of movement, although there is a substantial risk of intra-operative complications. Revision to a total shoulder replacement gives better results than to a hemiarthroplasty.

The results of primary total knee replacement performed on a group of haemophiliac patients in a single institution by the same surgeon using the same surgical technique and prosthesis are reported.

A total of 35 primary replacements in 30 patients were carried out between 1996 and 2005 and were reviewed retrospectively. The mean age of the patients was 31 years (24 to 42) and the mean follow-up was for 7.5 years (1 to 10). There were 25 patients with haemophilia A and five with haemophilia B. The HIV status and CD4 count were recorded, and Knee Society scores determined. Two patients had inhibitors to the deficient coagulation factor.

There were no early wound infections and only one late deep infection which required a two-stage revision arthroplasty, with a good final result. The incidence of infection in HIV-positive and negative patients was thus similar. One knee in a patient with inhibitor had excessive bleeding due to a pseudoaneurysm which required embolisation. The results were excellent in 27 knees (77%), good in six (17%) and fair in two (6%). The survival rate at 7.5 years taking removal of the prosthesis for loosening or infection as the end-point was 97%.

The mechanical survival of total knee replacements in haemophiliacs is very good. Our results confirm that this is a reproducible procedure in haemophilia, even in HIV-positive patients with a CD4 count > 200 mm³ and those with inhibitors. Our rate of infection was lower than previously reported. This could be due to better control of the HIV status with highly active anti-retroviral therapy and the use of antibiotic-loaded cement.

Glenoid replacement is technically challenging. Removal of a cemented glenoid component often results in a large osseous defect which makes the immediate introduction of a revision prosthesis almost impossible. We describe a two-stage revision procedure using a reversed shoulder prosthesis. Freeze-dried allograft with platelet-derived growth factor was used to fill the glenoid defect. Radiological incorporation of the allograft was seen and its consistency allowed the placement of a screwed glenoid component. There were no signs of new mature bone formation on histological examination.

The addition of platelet-derived growth factor to the allograft seems to contribute to an increase in incorporation and hardness, but does not promote the growth of new bone.

The results of proximal humeral replacement following trauma are substantially worse than for osteoarthritis or rheumatoid arthritis. The stable reattachment of the lesser and greater tuberosity fragments to the rotator cuff and the restoration of shoulder biomechanics are difficult. In 1992 we developed a prosthesis designed to improve fixation of the tuberosity fragments in comminuted fractures of the proximal humerus. The implant enables fixation of the fragments to the shaft of the prosthesis and the diaphyseal fragment using screws, washers and a special toothed plate. Between 1992 and 2003 we used this technique in 50 of 76 patients referred to our institution for shoulder reconstruction after trauma. In the remaining 26, reconstruction with a prosthesis and nonabsorbable sutures was performed, as the tuberosity fragments were too small and too severely damaged to allow the use of screws and the toothed plate.

The Constant score two years post-operatively was a mean of 12 points better in the acute trauma group and 11 points better in the late post-traumatic group than in the classical suture group.

We recommend this technique in patients where the tuberosity fragments are large enough to allow fixation with screws, washers and a toothed plate.

A consecutive series of 23 patients (25 ankles) with osteoarthritis of the ankle and severe varus or valgus deformity were treated by open arthrodesis using compression screws. Primary union was achieved in 24 ankles one required further surgery to obtain a solid fusion. The high level of satisfaction in this group of patients reinforces the view that open arthrodesis, as opposed to ankle replacement or arthroscopic arthrodesis, continues to be the treatment of choice when there is severe varus or valgus deformity associated with the arthritis.

Between January 1990 and December 2000 we carried out 226 SB Charité III disc replacements for lumbar disc degeneration in 160 patients. They were reviewed at a mean follow-up of 79 months (31 to 161) to determine the clinical and radiological outcome. The clinical results were collected by an independent observer, who was not involved in patient selection, treatment or follow-up, using a combination of outcome measures, including the Oswestry Disability Index. Pain was recorded using a visual analogue score, and the most recent radiographs were reviewed.

Survival of the device was analysed by the Kaplan-Meier method and showed a cumulative survival of 35% at 156 months when radiological failure was taken as the endpoint. The mean improvement in the Oswestry disability index scores after disc replacement was 14% (6% to 21%) and the mean improvement in the pain score was 1.6 (0.46 to 2.73), both falling below the clinically significant threshold. Removal of the implant was required in 12 patients, four because of implant failure.

These poor results indicate that further use of this implant is not justified.

Between 1993 and 2002, 58 GSB III total elbow replacements were implanted in 45 patients with rheumatoid arthritis by the same surgeon. At the most recent follow-up, five patients had died (five elbows) and six (nine elbows) had been lost to follow-up, leaving 44 total elbow replacements in 34 patients available for clinical and radiological review at a mean follow-up of 74 months (25 to 143). There were 26 women and eight men with a mean age at operation of 55.7 years (24 to 77).

At the latest follow-up, 31 excellent (70%), six good (14%), three fair (7%) and four poor (9%) results were noted according to the Mayo elbow performance score. Five humeral (11%) and one ulnar (2%) component were loose according to radiological criteria (type III or type IV). Of the 44 prostheses, two (5%) had been revised, one for type-IV humeral loosening after follow-up for ten years and one for fracture of the ulnar component. Seven elbows had post-operative dysfunction of the ulnar nerve, which was transient in five and permanent in two.

Despite an increased incidence of loosening with time, the GSB III prosthesis has given favourable mid-term results in patients with rheumatoid arthritis.

Between 1999 and 2005, 23 failed total ankle replacements were converted to arthrodeses. Three surgical techniques were used: tibiotalar arthrodesis with screw fixation, tibiotalocalcaneal arthrodesis with screw fixation, and tibiotalocalcaneal arthrodesis with an intramedullary nail. As experience was gained, the benefits and problems became apparent. Successful bony union was seen in 17 of the 23 ankles. The complication rate was higher in ankles where the loosening had caused extensive destruction of the body of the talus, usually in rheumatoid arthritis. In this situation we recommend tibiotalocalcaneal arthrodesis with an intramedullary nail. This technique can also be used when there is severe arthritic change in the subtalar joint. Arthrodesis of the tibiotalar joint alone using compression screws was generally possible in osteoarthritis because the destruction of the body of the talus was less extensive. Tibiotalocalcaneal arthrodesis fusion with compression screws has not been successful in our experience.

The aim of this biomechanical study was to investigate the role of the dorsal vertebral cortex in transpedicular screw fixation. Moss transpedicular screws were introduced into both pedicles of each vertebra in 25 human cadaver vertebrae. The dorsal vertebral cortex and subcortical bone corresponding to the entrance site of the screw were removed on one side and preserved on the other. Biomechanical testing showed that the mean peak pull-out strength for the inserted screws, following removal of the dorsal cortex, was 956.16 N. If the dorsal cortex was preserved, the mean peak pullout strength was 1295.64 N. The mean increase was 339.48 N (26.13%; p = 0.033). The bone mineral density correlated positively with peak pull-out strength.

Preservation of the dorsal vertebral cortex at the site of insertion of the screw offers a significant increase in peak pull-out strength. This may result from engagement by the final screw threads in the denser bone of the dorsal cortex and the underlying subcortical area. Every effort should be made to preserve the dorsal vertebral cortex during insertion of transpedicular screws.

We evaluated the long-term fixation of 64 press-fit cemented stems of constrained total knee prostheses in 32 young patients with primary malignant bone tumours. Initial stable fixation, especially in rotation, was achieved by precise fit of the stem into the reamed endosteum, before cementation. Complementary fixation, especially in migration and rotation, was obtained by pressurised antibiotic-loaded cement. The mean age at operation was 33 years (13 to 61). No patient was lost to follow-up; 13 patients died and the 19 survivors were examined at a mean follow-up of 12.5 years (4 to 21). Standard revision press-fit cemented stems were used on the side of the joint which was not involved with tumour (26 tibial and six femoral), on this side there was no loosening or osteolysis and stem survival was 100%. On the reconstruction side, custom-made press-fit stems were used and the survival rate, with any cause for revision as an end point, was 88%, but 97% for loosening or osteolysis. This longevity is similar to that achieved at 20 years with the Charnley-Kerboull primary total hip replacement with press-fit cemented femoral components.

We recommend this type of fixation when extensive reconstruction of the knee is required. It may also be suitable for older patients requiring revision of a total knee replacement or in difficult situations such as severe deformity and complex articular fractures.

We carried out a prospective randomised study designed to compare the penetration rate of acetabular polyethylene inserts of identical design but different levels of cross-linking at a minimum of four years follow-up. A total of 102 patients (102 hips) were randomised to receive either highly cross-linked Durasul, or contemporary Sulene polyethylene inserts at total hip replacement. A single blinded observer used the Martell system to assess penetration of the femoral head. At a mean follow-up of 4.9 years (4.2 to 6.1) the mean femoral head penetration rate was 0.025 mm/year (sd 0.128) in the Durasul group compared with 0.106 mm/year (sd 0.109) in the Sulene group (Mann-Whitney test, p = 0.0027). The mean volumetric penetration rate was 29.24 mm³/year (sd 44.08) in the Durasul group compared with 53.32 mm³/year (sd 48.68) in the Sulene group. The yearly volumetric penetration rate was 55% lower in the Durasul group (Mann-Whitney test, p = 0.0058). Longer term results are needed to investigate whether less osteolysis will occur.

We have evaluated the effect of vacuum aspiration of the iliac wing on the osseointegration of cement into the acetabulum. We entered a total of 40 patients undergoing primary total hip arthroplasty into two consecutive study groups. Group 1 underwent acetabular cement pressurisation for 60 seconds before insertion of the acetabular component. Group 2 had the same pressurisation with simultaneous vacuum suction of the ilium using an iliac-wing aspirator. Standard post-operative radiographs were reviewed blindly to assess the penetration of cement into the iliac wing. Penetration was significantly greater in the group with aspiration of the iliac wing.

We report the use of the reverse shoulder prosthesis in the revision of a failed shoulder hemiarthroplasty in 19 shoulders in 18 patients (7 men, 11 women) with severe pain and loss of function. The primary procedure had been undertaken for glenohumeral arthritis associated with severe rotator cuff deficiency.

Statistically significant improvements were seen in pain and functional outcome. After a mean follow-up of 44 months (24 to 89), mean forward flexion improved by 26.4° and mean abduction improved by 35°. There were six prosthesis-related complications in six shoulders (32%), five of which had severe bone loss of the glenoid, proximal humerus or both. Three shoulders (16%) had non-prosthesis related complications.

The use of the reverse shoulder prosthesis provides improvement in pain and function for patients with failure of a hemiarthroplasty for glenohumeral arthritis and rotator cuff deficiency. However, high rates of complications were associated with glenoid and proximal humeral bone loss.

Conventional amputation prostheses rely on the attachment of the socket to the stump, which may lead to soft-tissue complications. Intraosseous transcutaneous amputation prostheses (ITAPs) allow direct loading of the skeleton, but their success is limited by infection resulting from breaching of the skin at the interface with the implant. Keratinocytes provide the skin’s primary barrier function, while hemidesmosomes mediate their attachment to natural ITAP analogues. Keratinocytes must attach directly to the surface of the implant. We have assessed the proliferation, morphology and attachment of keratinocytes to four titaniumalloy surfaces in order to determine the optimal topography in vitro. We used immunolocalisation of adhesion complex components, scanning electron microscopy and transmission electron microscopy to assess cell parameters.

We have shown that the proliferation, morphology and attachment of keratinocytes are affected by the surface topography of the biomaterials used to support their growth. Smoother surfaces improved adhesion. We postulate that a smooth topography at the point of epithelium-ITAP contact could increase attachment in vivo, producing an effective barrier of infection.

We present a retrospective series of 170 cemented titanium straight-stem femoral components combined with two types of femoral head: cobalt-chromium (CoCr) alloy (114 heads) and alumina ceramic (50 heads). Of the study group, 55 patients (55 stems) had died and six (six stems) were lost to follow-up. At a mean of 13.1 years (3 to 15.3) 26 stems had been revised for aseptic loosening. The mean follow-up time for stable stems was 15.1 years (12.1 to 16.6).

Survival of the stem at 15 years was 75.4% (95% confidence interval (CI) 67.3 to 83.5) with aseptic failure (including radiological failure) as the end-point, irrespective of the nature of the head and the quality of the cement mantle. Survival of the stem at 15 years was 79.1% (95% CI 69.8 to 88.4) and 67.1% (95% CI 51.3 to 82.9) with the CoCr alloy and ceramic heads, respectively. The quality of the cement mantle was graded as a function of stem coverage: stems with complete tip coverage (type 1) had an 84.9% (95% CI 77.6 to 92.2) survival at 15 years, compared with those with a poor tip coverage (type 2) which had a survival of only 22.4% (95% CI 2.4 to 42.4). The poor quality of the cement mantle and the implantation of an alumina head substantially lowered the survival of the stem.

In our opinion, further use of the cemented titanium alloy straight-stem femoral components used in our series is undesirable.

We evaluated the use of a stemmed acetabular component in the treatment of severe acetabular deficiency during revision and complex primary total hip arthroplasty.

There were 31 hips of which 24 were revisions (20 for aseptic loosening, four for infection) and the remainder were complex primary arthroplasties. At a mean follow-up of 10.7 years (6 to 12.8), no component had been revised for aseptic loosening; one patient had undergone a revision of the polyethylene liner for wear. There was one failure because of infection. At the latest follow-up, the cumulative survival rate for aseptic loosening, with revision being the end-point, was 100%; for radiographic loosening it was 92% and for infection and radiographic loosening it was 88%. These results justify the continued use of this stemmed component for the reconstruction of severe acetabular deficiency.

We studied a consecutive series of 285 uncemented total hip replacements in 260 patients using the Taperloc femoral component and the T-Tap acetabular component. The outcome of every hip was determined in both living and deceased patients. A complete clinical and radiological follow-up was obtained for 209 hips in 188 living patients, followed for a mean of 14.5 years (10 to 18.9). They were divided into two groups, obese and non-obese, as determined by their body mass index. There were 100 total hip replacements in 89 patients in the obese cohort (body mass index ≥ 30 kg/m²), and 109 in 99 non-obese (body mass index < 30 kg/m²) patients. A subgroup analysis of 31 patients of normal weight (body mass index 20 kg/m² to 25 kg/m²) (33 hips) and 26 morbidly obese patients (body mass index ≥ 35 kg/m²) (30 hips) was also carried out.

In the obese group five femoral components (5%) were revised and one (1%) was loose by radiological criteria. Femoral cortical osteolysis was seen in eight hips (8%). The acetabular component was revised in 57 hips (57%) and a further 17 (17%) were loose. The mean Harris hip score improved from 52 (30 to 66) pre-operatively to 89 (49 to 100) at final follow-up. Peri-operative complications occurred in seven patients (7%).

In the non-obese group six (6%) femoral components were revised and one (1%) was loose. Femoral cortical osteolysis occurred in six hips (6%). The acetabular component was revised in 72 hips (66%) and a further 18 (17%) were loose. The mean Harris hip score increased from 53 (25 to 73) prior to surgery to 89 (53 to 100) at the time of each patient’s final follow-up radiograph.

No statistically significant difference was identified between the obese and non-obese patients with regards to clinical and radiological outcome or complications. The subgroup analysis of patients of normal weight and those who were morbidly obese showed no statistically significant difference in the rate of revision of either component.

Our findings suggest there is no evidence to support withholding total hip replacement from obese patients with arthritic hips on the grounds that their outcome will be less satisfactory than those who are not obese.

Although the use of constrained cemented arthroplasty to treat distal femoral fractures in elderly patients has some practical advantages over the use of techniques of fixation, concerns as to a high rate of loosening after implantation of these prostheses has raised doubts about their use. We evaluated the results of hinged total knee replacement in the treatment of 54 fractures in 52 patients with a mean age of 82 years (55 to 98), who were socially dependent and poorly mobile.

Within the first year after implantation 22 of the 54 patients had died, six had undergone a further operation and two required a revision of the prosthesis. The subsequent rate of further surgery and revision was low.

A constrained knee prosthesis offers a useful alternative treatment to internal fixation in selected elderly patients with these fractures, and has a high probability of surviving as long as the patient into whom it has been implanted.