Of 3000 patients diagnosed with primary malignant bone tumours and treated at our unit over the past 25 years, 234 (7.8%) were considered to be spindle cell sarcomas of bone (ie not osteosarcoma, chondrosarcoma, Ewing’s, chordoma or adamantinoma). We have analyzed their management and outcomes.

The diagnosis of these cases varied with fluctuations in the popularity of conditions such as MFH, fibrosarcoma and leiomyosarcoma with the passage of time. Treatment was with chemotherapy and surgery whenever possible. 36 patients had metastases at diagnosis and 17 had palliative treatment only because of age or infirmity. The most common site was the femur followed by the tibia, pelvis and humerus. The mean age was 45 and the mean tumour size 10.2cm at diagnosis. 25% of patients presented with a pathological fracture. Chemotherapy was used in 70% of patients the most common regime being cisplatin and doxorubicin. 35% of patients having neoadjuvant chemotherapy had a good (> 90% necrosis) response. The amputation rate was 22% and was higher in patients presenting with a fracture and in older patients not having chemotherapy.

With a mean follow up of 8 years the overall survival was 64% at 5 yrs and 58% at 10 yrs. Adverse prognostic factors included the need for amputation, older age and poor response to chemotherapy as well as a pathological fracture at presentation. The few patients with angiosarcoma fared badly but there was no difference in outcomes between patients with other diagnoses.

We conclude that patients with spindle cell sarcomas should be treated similarly to patients with osteosarcoma and can expect comparable outcomes. The histological diagnosis does not appear to predict behaviour.

Background: And Aims Pathological fractures of the proximal femur due to primary bone sarcomas are difficult to treat. The aim of the study was to assess the factors determining the outcomes following pathological fractures of the proximal femur due to primary bone sarcomas.

Methods: 93 patients with a pathological fracture of the proximal femur due to primary bone sarcomas were studied. The patient, tumour and treatment factors in relation to overall survival were analysed.

Results: There were 55 male and 38 female patients. The mean age was 47 years. The diagnoses were Chondrosarcoma -34, Osteosarcoma – 21, spindle cell sarcoma – 25, Ewing’s sarcoma -13. 74 patients had a pathological fracture at diagnosis and 19 patients had a fracture after the diagnosis. 17 patients had metastases at diagnosis. 24 patients had an intracapsular fracture. Limb salvage was possible in 60 patients (65%), 18 patients had an amputation and 15 patients had palliative treatment. 27% of the patients were referred after an unplanned surgery. The mean follow up was 49 months [range 0–302]. Twenty one patients [23%] had a local recurrence -10 patients had a diagnosis of chondrosarcoma, four patients had osteosarcoma and seven had spindle cell sarcoma. The overall five years survival was 37% [Ewing’s sarcoma 60%, Chondrosarcoma 57%, spindle cell sarcoma 28%, osteosarcoma 13% and dedifferentiated Chondrosarcoma 0% (p-0.002)]. Metastasis at diagnosis was a significant factor (p-0.04) affecting survival.

Conclusion: We conclude that a pathological fracture of the proximal femur due to osteosarcoma and dedifferentiated chondrosarcoma. carry a poor prognosis.

The two week wait has been established as a potential means of diagnosing malignant tumors earlier and thus hopefully leading to improvements in outcome. There remains controversy as to whether these clinic achieve this end or whether they just speed up treatment of patients already diagnosed (eg by imaging)

Aim: The aim of this study is to evaluate the diagnoses of all patient referred with a suspicious soft tissue lump to a two week wait clinic, to assess the diagnostic ‘hit rate’ and how many of these had already had imaging leading to a possible diagnosis of malignancy.

Method: Review of the diagnoses and referral criteria for all patients referred to a soft tissue sarcoma early diagnosis clinic.

Results: ** patients were referred under the 2 week wait criteria with a possible soft tissue sarcoma. ** (**%) turned out to have a malignant diagnosis. Of these, *8 were STS, ** were other soft tissue malignancies and ** were bone malignancices (** being….) Of the patients with a malignant diagnosis, only ** were referred directly to this hospital as a two week wait without previous investigation. Of the remainder *8 had undergone imag-9ing prior to referral leading to a suspicion of malignancy and ** were referred after investigation at another hospital. The average size of STS diagnosed after 2 week wait referral was ** which compared with **cm in those not referred via this route over the same time period.

Conclusion: Patients referred to a soft tissue 2 week wait clinic had a **% risk of malignancy. Of these, ** were new diagnoses suspected purely on clinical as opposed to imaging grounds.

Introduction: Malignant tumours of the foot and ankle are rare. The aim of the study was to document one of the largest series of malignant tumours affecting the foot and ankle and to assess the outcomes following limb salvage and amputation.

Methods: The study was a retrospective review of the patients with a malignant tumour of the foot and ankle. Demographic details, diagnosis, treatment and outcomes were retrieved from the electronic patient records containing information on over 20 000 patients seen over a 25 year period.

Results: Two hundred and twenty five patients had malignant tumours affecting the foot and ankle. It was common in the fifth decade (35 patients). The mean age was 46 years. The commonest diagnosis was synovial sarcoma (40 patients) followed by chondrosarcoma (23 patients) and Ewing’s sarcoma (21 patients). The mean tumour size was 5.6 cm (0.8 to 17.5 cm). 82 patients (37%) underwent an unplanned excision and 13% (29 patients) presented with metastases at diagnosis. Primary bone tumours were 28% (64 patients), soft tissue sarcomas were 62% and metastatic tumours were 8% while lymphoreticular malignancies were (1%). Limb salvage was possible in 71% (156 patients). 29% (65 patients) had a below knee amputation. 7% (15 patients) had a local recurrence. The 5 years survival was 63%. The 5 years survival for the patients who had limb salvage was 68% compared with 54% for the patients who had an amputation (p 0.03).

Conclusion: Though amputation can provide better local control, limb salvage surgery improves survival.

Low grade central osteosarcoma is a rare intramedullary bone producing tumour. It accounts for only 1–2% of all osteosarcomas. Due to the indolent nature of low grade central osteosarcoma, achieving a correct and prompt diagnosis is the real challenge both from imaging and histology, particularly as it may resemble a benign condition, i.e. Fibrous Dysplasia.

We have reviewed 15 cases of low grade central osteosarcoma with long term follow-up (2 to 22 years) to identify problems in diagnosis and treatment and to assess outcome.

There were 7 females and 8 males with a mean age of 37 yrs (range 11 to 72 years); 13 cases arose in the lower limb (8 femur, 4 tibia, 1 os calcis), 1 in the pelvis and 1 in the upper limb. The average duration of symptoms prior to presentation was over 2yrs. A primary diagnosis of low grade central osteosarcoma was achieved for only 6 cases (4 open and 2 needle biopsies), in the other 9 the primary diagnoses were GCT, cystic lesion or fibrous lesion (both benign and malignant) and all of them had undergone treatment (usually curettage with or without bone grafting for this). Definitive treatment was with surgery attempting to obtain wide margins. Marginal excision was associated with local recurrence in three cases but there were no local recurrences in patients who had a wide excision, even in those with prior treatment. Only one patient has died following the development of multiple metastases after 9 years. The survival rate is 90% at 15 years.

We present this study to show the difficulties in diagnosing this rare type of osteosarcoma and to highlight the importance of wide surgical margins to obtain local control.

Aim: To investigate the effectiveness of silver coated titanium prostheses in preventing periprosthetic infection in a group of very high risk patients.

Methods: Periprosthetic infection is one of the main problems in limb salvage surgery, especially for tibial and pelvic prostheses or following revision surgery, particularly if it has been done for a previous infection. We have used silver coated prostheses in 21 cases on a named patient basis and have now assessed the results.

Results: Between July 2006 and June 2008 21 patients had a silver coated prosthesis inserted. 11 patients were having a second stage revision after a previous infection, 6 were having a pelvic prosthesis inserted, 3 were having a primary tibial replacement and one a one stage proximal tibial revision. Three patients developed a postoperative infection, two of the pelvic replacements and one infected revision (a total femur replacement). Of these only one patient required removal of the prosthesis (for overwhelming coliform infection in a pelvic replacement) whilst the other two infections both settled with antibiotics and washout.

Discussion: The anticipated risk of infection in this high risk group would have been around 20%. The actual infection rate was 14% but two of the infections completely resolved with relatively modest treatment. This suggests that the silver coating may not only have a role in preventing infection but also enhancing control of infection should it arise. There were no other side effects and we believe that these preliminary results are encouraging and should lead to a further evaluation of silver for preventing infection around prostheses.

Aim: To review our experience of managing patients with aneurysmal bone cysts (ABC)

Method: We reviewed the medical records and radiographs of all patienst with aneurismal bone cyst treated at our unit over a 25 year period. During that time the policy of the unit was to treat ABCs with biopsy/curettage without use of adjuvants or bone grafting. Patients were followed up with regular Xrays until healing had taken place. Local recurrences were again treated with curettage, occasionally supplemented with embolisation or bone grafting.

Results: 237 patients (128 female, 109 male), with a median age of 14 yrs (range 1 to 76) received treatment. The cyst size varied from 1 to 20 cm and the median duration of symptoms was 16 weeks (range 0 to 8 yrs). The most common sites were the tibia (55) followed by the femur (41) then the pelvis (29) and humerus (27). 35 (15%) of the patients presented with a pathological fracture. Primary treatment was by curettage alone in 195, curettage and bone grafting in 7, aspiration and injection of steroids or bone marrow in 7, excision in 5 and observation alone in 17. The rate of local recurrence requiring further surgery was 12% with all local recurrences (bar one) arising within 18 months. Local recurrence was not related to site, age, sex or whether the patient had previous treatment or not. Local recurrences were managed with curettage alone in 19 of the 23 cases, with one having embolisation, one excision and 2 curettage and bone grafting. This was successful in all but 3 cases who were controlled with a third procedure.

Conclusion: The local control rate of ABCs with simple curettage is 88% which is as good as those published for any other technique. We recommend biopsy in all cases and limited curettage at the same time, many ABCs will heal with this simple procedure. Full curettage is needed for those showing no signs of healing within 4 weeks. Local recurrence is very unusual after 18 months.

Introduction: Parosteal osteosarcoma is a rare, mostly low grade tumor arising from the surface of the bone. The aim of the study was to establish risk factors for local recurrence (LR) and death.

Method: Retrospective review of a prospectively collected database with complete up-to-date follow up.

Results: 63 patients were diagnosed and treated at our centre from 1978 to 2007. The most common site was the femur (37) followed by the tibia (10) and humerus (9). The median size was 7cm (2 – 16). At the time of diagnosis 27 (43%) were low grade, 13 (21%) were low grade but had invaded the medullary canal and 8 (13%) had a high grade component; the data was missing in 15 patients (24%). All patients were operated.

LR arose in 11 patients (17.5%) at a mean of 38 months and was related to surgical margins. There were no LR in the 26 wide resections, but the risk of LR was 20% in those with a marginal and 46% in those with an intralesional excision. LR appeared as a higher grade than the primary tumor in 3 patients (27%).

9 (14.3%) patients developed metastases at a median of 22 months (6 – 123). Five patients developed both LR and metastases. Seven patients died from the tumor, 2 patients are alive with metastatic disease.

None of the patients with a low grade tumor and wide or marginal margins died of tumor, but 2 patients (18%) operated intralesional died of tumor.

Discussion: There was a statistically significant relationship between surgical margins and LR (p< 0.001). All patients with low grade parosteal osteosarcoma operated with wide margins have been cured. LR appeared often in higher grade than the primary tumor. Patients with low grade tumor and intralesional margins had 18% risk to tumor-ralated death.

Ewing’s sarcoma principally arises in bone but can also present as a soft tissue tumour. Very few studies have assessed the outcomes of extra-skeletal Ewing’s sarcomas. This study compares the oncological outcomes of the two forms of Ewing’s sarcomas to see if there is any difference in prognostic factors.

198 patients with primary, non metastatic Ewing’s sarcoma diagnosed between 1980 and 2005 were identified from our database. There were 118 males and 80 females with a median age of 15 years. The three most common sites of diagnosis were the femur (24%), pelvis (15%) and tibia (13%). There were 169(85%) bony Ewing’s and 29 (15%) extra-skeletal Ewing’s sarcomas. All patients received chemotherapy. 86% of the patients had surgery for local control but 28(14%) patients had radiotherapy.

The overall survival at five years was 89% and was related to the age of patient (92% < 16years p=0.005), size (p=0.03) and site of tumour (p=0.004) as well as the response to chemotherapy. There was no difference in the overall survival of patients with bony Ewing’s (90%) and extra-skeletal Ewing’s (85%) (p=0.85). There was a 10% risk of local recurrence at 5 years with site of tumour (p=0.01) and surgical excision (p=0.05) being significant prognostic factors. The risk of local recurrence was also not related to the type of Ewing’s sarcoma.

This large series has shown that the oncological outcomes of Ewing’s sarcoma is related to tumour characteristics, patient age and treatment factors and not determined by the tissue component.

Aim: To identify the clinical features of patients who present with soft tissue metastases (from a cancer elsewhere) and to identify the source of the cancer.

Method: In a database containing details of 7242 patients referred to our unit for investigation of a soft tissue lump, only 100 of these patients were found to have a soft tissue metastasis (1.4%). We analysed their presenting features and identified the site of the primary malignancy.

Results: The most common presentation of soft tissue metastases was a painless lump The lumps ranging from 2 to 35 cm (mean 8.3cm) with 78% of the lumps located deep to the fascia. The mean age at presentation was 62 and there were equal males and females. 53 had a past history of malignancy. Of these 53, 52 had metastases from the same primary (lung 11, melanoma 10, kidney 9, GI tract 4, breast 6, bladder 4 and others in 9). The other 47 patients had no past history of malignancy and the metastasis was the first presentation of malignancy. The primary sites in these cases were: lung in 19, GI tract 4, kidney 2, melanoma 9, other 3 and unknown (despite investigations) in 10. There was no correlation of the site of the metastases with the primary tumour.

Of the 7242 patients with soft tissue lumps, 476 had a past history of malignancy. Of these patients, only 12% actually had a soft tissue metastasis while 28% had a benign diagnosis, 55% a soft tissue sarcoma and 5% other malignancy.

Conclusion: Patients with a past medical history of malignancy and a new soft tissue lump have a 12% chance of it being a soft tissue metastasis. If a lump proves to be a soft tissue metasasis, the lung is the most common primary site.

Introduction: The population of the UK is getting older. Patients over the age of 80 (the older old) are increasingly presenting with musculoskeletal tumours that require major surgery. We have investigated the success or otherwise of endoprosthetic replacements after tumour excision in this population.

Methods: We looked at an oncological database to identify patients over the age of 80 who had an endprosthetic replacement after tumour excision. We reviewed the records of all patients over the age of 80 who had an endoprosthetic replacement to assess the oncological and functional outcomes of the procedure.

Results: 17 patients over the age of 80 had an endoprosthesis over the past 10 years. The main indication was for metastatic disease (9 patients) but 6 had primary malignant bone tumours. The most common site was the distal femur in 9 followed by the proximal femur in 5. Most of the patients had associated co-morbidity (12 were ASA 2 or 3). There were no perioperative deaths but 2 patients had early complications with one having a paralytic ileus and one a chest infection. The median survival of the patients was 2 years with death being due to progressive metastases in most. 3 had late complications to do with the prosthesis including one infection in a proximal tibial replacement and a late dislocation at 2 years in a proximal femoral replacement. Two patients developed local recurrence treated by local excision and radiotherapy. There were no amputations or revisions.

Conclusions: Endoprosthetic replacements have a useful role to play in the surgical management of elderly patients. Although they have significant comorbidity most do well. Functional results are less good than in the younger population but most patients regain their independence and are free of pain.

About one third of patients who require one knee replacement have significant bilateral symptoms and will require surgery on both knees before achieving their full functional potential. The options for these patients are either to have one-stage bilateral knee replacements or two-stage knee replacements. Our aim was to compare the relative local and systematic morbidity of patients who had one-stage bilateral knee arthroplasty with those of patients who had unilateral total knee arthroplasty in a retrospective, consecutive cohort of patients to evaluate the safety of one-stage bilateral total knee arthroplasty. Seventy-two patients treated with one-stage bilateral knee replacements were matched for age, gender and year of surgery with 144 patients who underwent unilateral knee arthroplasty. We found one-stage bilateral arthroplasty was associated with significantly increased risks of wound infection, deep infection, cardiac complications and respiratory complications compared to unilateral knee arthroplasty.

No increased risk of thromboembolic complications or mortality was found.

We conclude that one-stage bilateral total knee arthroplasty is associated with increased risk of both systematic and local complications compared with unilateral knee replacement and therefore should be performed on only selective cases.

Our centre has used a specially designed custom-made endoprostheses with curved stems to reconstruct femoral defects in patients with residual short proximal femur after excision of primary bone sarcoma over the last 18 years. Two designs of endoprostheses with curved intramedullary stems were used: the rhinohorn stem type and the bifid stem type. We report the safety, survival and functional outcome of this form of reconstruction.

Twenty six patients who had these special endoprosthesis reconstruction were studied. The median age was 16 years (range 7 to 60 years). Prostheses with rhino horn stems were used in 15 patients and bifid-stem in 1 1 patients. Twenty patients had the prostheses inserted as a primary procedure after excision of primary bone sarcoma, and in six patients the prostheses were inserted after revision surgery of failed distal femur endoprostheses. Seventeen patients (65%) were alive and free of disease at a median follow-up of 98 months (12 to 203 months) and nine patients had died of metastatic disease. Local recurrence developed in two patients (1 0%) out of the 20 patients. Surgical complications occurred in five patients (191/o). Deep infections occurred in two patients (8%) requiring revision surgery in one patient. Prosthetic failure, occurred in nine patients (35%). The cumulative survival of prostheses was 69% at five years and 43% at 10 years. Musculoskeletal Tumour Society mean functional score was 83% (53% to 97%).

In conclusion, preservation of a short segment of the proximal femur and the use of endoprostheses with curved stems for reconstruction of the femur is technically possible. There is an increased risk of fracture of the prostheses decreasing the survival rate. Functional outcome of patients with this form of reconstruction is not significantly different from the functional outcome of patients who have proximal femur or total femur endoprosthetic reconstruction. This operation is particularly desirable in skeletally immature patients and allows normal development of the acetabulum.

The purpose of this retrospective study was to analyze the risk factors, causes, bacteriology of deep infection following extensible endoprosthetic replacement for bone tumours in children and to review our experience in the treatment of 20 patients with infected prostheses.

123 patients with extensible endoprostheses were treated between 1983 and 1998. Three types of prostheses, which differed in the lengthening mechanism used, were implanted. 20 of these were diagnosed to have deep infection. Patients were divided into 3 groups: group I 5 patients were treated with a single stage revision, group 11– 13 patients were treated with a two stage revision procedure, group Ill- 2 patients had a primary amputation. Control of infection was assessed clinically and with inflammatory markers. Function was assessed using the MSTS score.

The overall incidence of infection was 16%. The incidence of infection at the proximal tibia and distal femur was 27% and 14% respectively. Staphylococcus epidermidis was the most common organism. The most common clinical features were pain and swelling around the prostheses. Infection in most cases was immediately preceded by an operative procedure or by distant a focus of infection. The number of operative procedures and the site of the prosthesis were significant risk factors. The success rate was 20% in Group 1 and 84% in Group II. Amputation was the salvage procedure of choice for failed revision procedures. The mean MSTS functional score was 83% in patients in whom the infection was controlled.

The incidence of deep infection is high following extensible endoprostheses. The site of the prosthesis and the number of operative procedures are significant risk factors. The type of prosthesis used is not a risk factor. Two-stage revision is successful in controlling infection in a majority of these cases.

This is a retrospective study of 70 patients with chondroblastoma treated between 1973 to 2000. Of these 70 patients, 53 had their primary procedure performed at our unit in the form of an intralesional curettage. The purpose of this study was to determine the rates of recurrence and the functional outcomes following this technique. Factors associated with aggressive tumour behaviour were also analysed. The patients were followed up for at least 22 months, up to a maximum of 27 years. 6 out of these 53 cases (11. 3%) had a histologically proven local recurrence. Three patients underwent a second intralesional curettage procedure and had no further recurrences. Two patients had endoprosthetic replacement of the proximal humerus and one patient underwent a below knee amputation following aggressive local recurrences. One patient had the rare malignant metastatic chondroblastoma and died eventually. The mean MSTS score was 94. 1%. We conclude that meticulous primary intralesional curettage without any additional procedure can achieve low rates of local recurrence and excellent long-term functional results.

Malignant peripheral nerve sheath tumours (MPNSTs) constitute 10% of soft tissue sarcomas. A significant proportion arise in neurofibromatosis type 1 (NF1). Several publications have compared MPNST survival in sporadic and NF1 patients, without consensus on whether NF1 is an independent factor for poor prognosis.

Clinical and histological data from 135 proven MPNSTs were analysed from 2 national centres for soft tissue tumour surgery diagnosed from 1979 to 2000. 129 patients had follow-up data from 6 months to 21 years. 35 were from patients with NF1. Local treatment involved surgery in surgery in 95%, radiotherapy in 44% and chemotherapy in 21%.

NF1 patients were younger than those with sporadic tumours (median age 26 years vs 53 years, p< 0. 001). Overall MPNST survival was almost identical to that in soft tissue sarcomas as a whole, but was worse in NF1 than in sporadic tumours (33% vs 72% at 30 months [p< 0. 01], 17% vs 39% at 60 months, 6% vs 21% at 120 months). A trend towards shorter time to local recurrence was seen in NF1, but not time to metastasis. Superficial tumours gave improved prognosis. Tumour volume over 100ml was associated with worse survival (46% vs 91% at 30 months, p< 0. 02), as was histological grade (80% high grade vs 25% low grade at 60 months, p< 0. 01). In terms of location, a non-significant over-representation of NF1 MPNSTs in the sciatic and brachial plexii was identified.

NF1 and sporadic MPNSTs exhibited no difference in depth or tumour volume profile, although NF1 tended towards higher grade. Analysis of survival in only high grade tumours, however, still resulted in a significant survival disadvantage in NF1 (33% vs 70% at 30 months, p< 0. 01). Removal of brachial and sciatic plexus tumours from analysis did not affect survivorship profiles in NF1 and sporadic MPNSTs.

Grade, volume and tumour depth correlate with survival; only 7 of 45 patients with deep high grade tumours over 100ml volume were observed to survive beyond 2 years. MPNST survival is worse in NF1 than sporadic tumours. Grade, depth, site and volume differences could not explain this disadvantage.

Few studies of wound complications following limb salvage surgery for soft tissue sarcomas separate anatomical compartments. Forty-nine patients with adductor compartment sarcomas underwent limb salvage surgery, 43% developing significant wound complications, 25% requiring further surgery and 20% had delays in adjuvant radiotherapy as a result. Prior surgery by non tumour surgeons and previous radiotherapy led to an increased risk of wound healing problems. In this particular group of patients, special attention should be made to prevent wound healing complications, possibly involving plastic surgeons at an earlier stage of management.

Reconstruction of the shoulder joint following resection of the proximal humerus for bone tumours remains controversial. We report the long term functional results of the simplest form of reconstruction – an endoprosthesis.

One hundred patients underwent endoprosthetic replacement of the proximal humerus between 1976 and 1998. Thirty eight had osteosarcoma, 17 had chondrosarcoma, 16 had metastases and 9 had Ewing’s sarcoma. Mean age was 36 years (range 10 to 80 yrs). Survivorship of patients and prostheses were calculated. Function was assessed using the Musculoskeletal tumour society (MSTS) and Toronto extremity salvage (TESS) scoring systems. Thirty patients could come to the clinics for MSTS scoring and 38 out of 49 alive patients replied to the TESS questionnaires sent out to them.

The overall survival of the patients was 42% at 10 years. Local recurrence (LR) arose in 16 patients, being most common in chondrosarcoma (26%) and osteosarcoma (22%) and arose in 50% of patients with these tumours who had marginal excisions. Of these 16 patients, 8 had forequarter amputations whilst the remainder had excisions and radiotherapy. Mean time to LR was 12 months and all but two of these 16 patients subsequently died within a mean of 18 months.

The prostheses proved reliable and dependable. Only 9 required further surgery of any sort, 2 needing minor surgery to correct subluxation and 7 needing revisions, one for infection after radiotherapy and six for loosening – three after trauma. The survivorship of the prosthesis without any further surgery was 86. 5% at 20years. The survivorship of the limb without amputation was 93% at 20 years.

The functional outcome was very predictable. Most patients had only 45 degrees of abduction although three patients had normal movements. The mean MSTS functional score was 79% and the mean TESS score was also 79%. There was a high level of patient satisfaction but difficulty was encountered especially in lifting and in all activities above shoulder height. Endoprosthetic replacement of the proximal humerus is a predictable procedure providing reasonable function of the arm below shoulder height. The endoprostheses have proved highly dependable with a low re-operation rate. There is a high risk of local recurrence after inadequate surgery which should be avoided if possible.

This retrospective clinical study describes our experience of the use of growing endoprostheses in children with primary malignant tumours of the proximal femur and analyses the results.

Between 1983 and 1996 we treated nine children with primary bone tumors of the proximal femur by resection and proximal femoral extensible replacements. Outcomes measured were function of the limb using Musculoskeletal Tumor Society score, oncologic outcome, complications and equalization of limb length. Results: Four patients died as a result of pulmonary metastases. The remaining five patients were observed for an average follow-up period of 7. 6 years (range 11–12. 7 years). One patient had a hindquarter amputation for uncontrolled infection. In these five patients we performed an average of 10. 2 operative procedures per patient (range of 3–17 procedures) including 5 lengthening procedures (range of 1–8 procedures) and a mean total extension of 69. 7 mm per patient. Acetabular loosening and hip dislocations were the most frequent complications. Only two patients have not had a revision or a major complication. Despite this, 4 children are alive with a functioning lower limb and a mean Musculoskeletal Tumour Society functional score of 77. 6%. The limb length discrepancy was less than 1 0 mm in three of these patients. The remaining patient has a discrepancy of 50 mm and is awaiting further limb equalization procedures.

Extendible endoprostheses of the proximal femur in selected children is a viable reconstructive procedure. It allows for equalization of limb length and the ability to walk without the use of mobility aids.

Between 1982 and 1997, twenty-six children between the age of 2 and 15 (mean age 10. 6 years) underwent proximal femoral replacement. Twenty have survived and all but three have reached skeletal maturity.

Sequential radiographs have been reviewed with particular reference to acetabular development and fixation of the prostheses. Initially a cemented acetabular component was inserted, but recently uncemented implants and unipolar femoral heads that exactly fit the acetabulum have been used.

In older children the acetabulum develops normally and the components remain well fixed. One of nine children over thirteen years with a cemented acetabulum needed revision for loosening and one suffered recurrent dislocations.

In younger children the acetabulum continues to develop at the triradiate cartilage, so a cemented acetabulum grows away from the ischiopubic bar. As the component is fixed proximally, it becomes increasingly vertical and will almost inevitably loosen. In our study six of eight children under 13 years of age with a cemented acetabulum needed revision for loosening.

Unipolar replacements in younger children tend to erode the superior acetabular margin. Femoral head cover is difficult to maintain, and of four unipolar implants in children under thirteen, two required acetabular augmentation.

Cemented cups may be unsuitable for children under thirteen years but our results are not statistically significant. In this age group, unipolar implants may be more appropriate but they have serious potential complications. In children over thirteen, cemented implants survive longer. The number of uncemented implants in our study is too small to comment on long-term survival.