Aims. Following the resection of an extensive amount of bone in the treatment of a tumour, the residual segment may be insufficient to accept a standard length intramedullary cemented stem. Short-stemmed endoprostheses conceivably have an increased risk of aseptic loosening. Extra-cortical plates have been added to minimise this risk by supplementing fixation. The aim of this study was to investigate the survivorship of short-stemmed endoprostheses and extra-cortical plates. Patients and Methods. The study involved 37 patients who underwent limb salvage surgery for a primary neoplasm of bone between 1998 and 2013. Endoprosthetic replacement involved the proximal humerus in nine, the proximal femur in nine, the distal femur in 13 and the proximal tibia in six patients. There were 12 primary (32%) and 25 revision procedures (68%). Implant survivorship was compared with matched controls. The amount of bone that was resected was > 70% of its length and statistically greater than the standard control group at each anatomical site. Results. The mean follow-up was seven years (one to 17). The mean length of the stem was 33 mm (20 to 60) in the humerus and 79 mm (34 to 100) in the lower limb. Kaplan-Meier analysis of survival of the implant according to anatomical site confirmed that there was no statistically significant difference between the short-stemmed endoprostheses and the standard stemmed controls at the proximal humeral (p = 0.84), proximal femoral (p = 0.57), distal femoral (p = 0.21) and proximal tibial (p = 0.61) sites. In the short-stemmed group, no implants with extra-cortical plate osseointegration suffered loosening at a mean of 8.5 years (range 2 to 16 years). Three of ten (30%) without osseointegration suffered aseptic loosening at a mean of 7.7 years (range 2 to 11.5 years). Conclusion. When extensive resections of bone are required in the surgical management of tumours, and in revision cases, the addition of extra-cortical plates to short medullary stems has shown non-inferiority to standard length medullary stems and minimises aseptic failure. Cite this article: Bone Joint J 2017;99-B:1689–95

Aims. Endoprosthetic reconstruction following distal femur tumour resection has been widely advocated. In this paper, we present the design of an uncemented endoprosthesis system featuring a short, curved stem, with the goal of enhancing long-term survivorship and functional outcomes. Methods. This study involved patients who underwent implantation of an uncemented distal femoral endoprosthesis with a short and curved stem between 2014 and 2019. Functional outcomes were assessed using the 1993 version of the Musculoskeletal Tumour Society (MSTS-93) score. Additionally, we quantified five types of complications and assessed osseointegration radiologically. The survivorship of the endoprosthesis was evaluated according to two endpoints. A total of 134 patients with a median age of 26 years (IQR 16 to 41) were included in our study. The median follow-up time was 61 months (IQR 56 to 76), and the median functional MSTS-93 was 83% (IQR 73 to 91) postoperatively. Results. Overall, 21 patients (16%) encountered complications, and the rate of aseptic loosening was 7% (9/134). The survival rate up to 8.5 years was 93% for aseptic loosening as the endpoint, and 88% for any reason as the endpoint, retrospectively. Conclusion. The use of an uncemented distal femoral endoprosthesis with a short, curved stem demonstrated a low incidence of aseptic loosening and achieved long-term survivorship of up to nine years. Meanwhile, aseptic loosening typically occurs in the early stage postoperatively. Cite this article: Bone Joint J 2024;106-B(9):1000–1007

Aims. Hydroxyapatite (HA)-coated collars have been shown to reduce aseptic loosening of massive endoprostheses following primary surgery. Limited information exists about their effectiveness in revision surgery. The aim of this study was to radiologically assess osteointegration to HA-coated collars of cemented massive endoprostheses following revision surgery. Methods. Retrospective review of osseointegration frequency, pattern, and timing to a specific HA-coated collar on massive endoprostheses used in revision surgery at our tertiary referral centre between 2010 to 2017 was undertaken. Osseointegration was radiologically classified on cases with a minimum follow-up of six months. Results. In all, 39 patients underwent radiological review at mean 43.5 months; 22/39 (56.4%) showed no osseointegration to the collar. Revision endoprostheses for aseptic loosening were less likely to show osseointegration compared with other indications for revision. Oncological cases with previous or current infection were more likely to show osseointegration to ≥ 1 collar side than those without evidence of prior infection. Conclusion. This seven-year review identified osseointegration of HA-coated collars after revision surgery is less likely (43.6%, 17/39) than after primary surgery. Young patients who undergo revision surgery following initial oncological indication may benefit the most from this collar design. Use in revision oncological cases with a history of infection may be beneficial. HA-coated collars showed limited benefit for patients undergoing revision for failed arthroplasty with history of infection. Cite this article: Bone Jt Open 2021;2(6):371–379

Aims. Rotating-hinge knee prostheses are commonly used to reconstruct the distal femur after resection of a tumour, despite the projected long-term burden of reoperation due to complications. Few studies have examined the factors that influence their failure and none, to our knowledge, have used competing risk models to do so. The purpose of this study was to determine the risk factors for failure of a rotating-hinge knee distal femoral arthroplasty using the Fine-Gray competing risk model. Methods. We retrospectively reviewed 209 consecutive patients who, between 1991 and 2016, had undergone resection of the distal femur for tumour and reconstruction using a rotating-hinge knee prosthesis. The study endpoint was failure of the prosthesis, defined as removal of the femoral component, the tibial component, or the bone-implant fixation; major revision (exchange of the femoral component, tibial component, or the bone-implant fixation); or amputation. Results. Multivariate Fine-Gray regression analyses revealed different hazards for each Henderson failure mode: percentage of femoral resection (p = 0.001) and extent of quadriceps muscle resection (p = 0.005) for overall prosthetic failure; extent of quadriceps muscle resection (p = 0.002) and fixation of femoral component (p = 0.011) for type 2 failure (aseptic loosening); age (p = 0.009) and percentage of femoral resection (p = 0.019) for type 3 failure (mechanical failure); and type of joint resection (p = 0.037) for type 4 (infection) were independent predictors. A bone stem ratio of > 2.5 reliably predicted aseptic loosening. Conclusion. We identified independent risk factors for overall and cause-specific prosthetic failure after rotating-hinge knee distal femoral arthroplasty using a competing risk Fine-Gray model. A bone stem ratio > 2.5 reliably predicts aseptic loosening. An accurate knowledge of the risks of distal femoral arthroplasty after resection for tumour assists surgical planning and managing patient expectations. Cite this article: Bone Joint J 2021;103-B(8):1405–1413

Aims. Aseptic loosening is a major cause of failure in cemented endoprosthetic reconstructions. This paper presents the long-term outcomes of a custom-designed cross-pin fixation construct designed to minimize rotational stress and subsequent aseptic loosening in selected patients. The paper will also examine the long-term survivorship and modes of failure when using this technique. Patients and Methods. A review of 658 consecutive, prospectively collected cemented endoprosthetic reconstructions for oncological diagnoses at a single centre between 1980 and 2017 was performed. A total of 51 patients were identified with 56 endoprosthetic implants with cross-pin fixation, 21 of which were implanted following primary resection of tumour. Locations included distal femoral (n = 36), proximal femoral (n = 7), intercalary (n = 6), proximal humeral (n = 3), proximal tibial (n = 3), and distal humeral (n = 1). Results. The median follow-up was 132 months (interquartile range (IQR) 44 to 189). In all, 20 stems required revision: eight for infection, five for structural failure, five for aseptic loosening, and two for tumour progression. Mechanical survivorship at five, ten, and 15 years was 84%, 78%, and 78%, respectively. Mechanical failure rate varied by location, with no mechanical failures of proximal femoral constructs and distal femoral survivorship of 82%, 77%, and 77% at five, ten, and 15 years. The survivorship of primary constructs at five years was 74%, with no failure after 40 months, while the survivorship for revision constructs was 89%, 80%, and 80% at five, ten, and 15 years. . Conclusion. The rate of mechanical survivorship in our series is similar to those reported for other methods of reconstruction for short diaphyseal segments, such as compressive osseointegration. The mechanical failure rate differed by location, while there was no substantial difference in long-term survival between primary and revision reconstructions. Overall, custom cross-pin fixation is a viable option for endoprosthetic reconstruction of short metaphyseal segments with an acceptable rate of mechanical failure. Cite this article: Bone Joint J 2019;101-B:724–731

Aims. The aim of this study was to investigate the feasibility of application of a 3D-printed megaprosthesis with hemiarthroplasty design for defects of the distal humerus or proximal ulna following tumour resection. Methods. From June 2018 to January 2020, 13 patients with aggressive or malignant tumours involving the distal humerus (n = 8) or proximal ulna (n = 5) were treated by en bloc resection and reconstruction with a 3D-printed megaprosthesis with hemiarthroplasty, designed in our centre. In this paper, we summarize the baseline and operative data, oncological outcome, complication profiles, and functional status of these patients. Results. Preparation of the prosthesis was a mean of 8.0 days (SD 1.5), during which time no patients experienced tumour progression. The mean operating time and intraoperative blood loss were 158.1 minutes (SD 67.6) and 176.9 ml (SD 187.8), respectively. All of the prostheses were implanted successfully. During a mean follow-up of 25.7 months (SD 7.8), no patients died, but four had complications (two superficial wound problems, one temporary palsy of radial nerve, and one dislocation). No aseptic loosening, structural failure, infection, heterotopic ossification, or degenerative arthritis was seen in this study. The mean flexion of the elbow was 119.6° (SD 15.9°) and the mean extension lag was 11.9° (SD 13.8°). The mean Musculoskeletal Tumor Society 93 score and Mayo Elbow Performance Score were 28.4 (SD 0.9) and 97.7 (SD 4.4), respectively. Conclusion. The custom-made, 3D-printed megaprosthesis with hemiarthroplasty is a feasible option for functional reconstruction after resection of a tumour in the distal humerus or proximal ulna. Cite this article: Bone Joint J 2022;104-B(6):747–757

Aims. For rare cases when a tumour infiltrates into the hip joint, extra-articular resection is required to obtain a safe margin. Endoprosthetic reconstruction following tumour resection can effectively ensure local control and improve postoperative function. However, maximizing bone preservation without compromising surgical margin remains a challenge for surgeons due to the complexity of the procedure. The purpose of the current study was to report clinical outcomes of patients who underwent extra-articular resection of the hip joint using a custom-made osteotomy guide and 3D-printed endoprosthesis. Methods. We reviewed 15 patients over a five-year period (January 2017 to December 2022) who had undergone extra-articular resection of the hip joint due to malignant tumour using a custom-made osteotomy guide and 3D-printed endoprosthesis. Each of the 15 patients had a single lesion, with six originating from the acetabulum side and nine from the proximal femur. All patients had their posterior column preserved according to the surgical plan. Results. Postoperative pathological assessment revealed a negative surgical margin was achieved in all patients. At final follow-up, 13.3% (2/15) died and no recurrence occurred. The overall survival was 81.7% at five years. None of the patients showed any signs of aseptic loosening, and no wound healing issues were observed. In total, 20% (3/15) developed complications, with two cases of early hip dislocation and one case of deep infection. The cumulative incidence of mechanical and non-mechanical failure in this series was 13.7% and 9.3%, respectively, at five years. In this cohort, the mean time to full weightbearing was 5.89 (SD 0.92) weeks and the mean Musculoskeletal Tumor Society score was 24.1 (SD 4.4). Conclusion. For patients with a hip joint tumour who met the inclusion criteria and were deemed suitable for posterior column preservation, a custom-made osteotomy guide combined with 3D-printed endoprosthesis is worth performing when treating patients who require extra-articular resection of the hip joint, as it can achieve adequate margin for local control, maximize bone preservation to maintain pelvic ring integrity, reduce the risk of complications by simplifying the surgical procedure, and allow for more precise reconstruction for better function. Cite this article: Bone Jt Open 2024;5(11):1027–1036

Aims. Limb salvage in bone tumour patients replaces the bone with massive segmental prostheses where achieving bone integration at the shoulder of the implant through extracortical bone growth has been shown to prevent loosening. This study investigates the effect of multidrug chemotherapy on extracortical bone growth and early radiological signs of aseptic loosening in patients with massive distal femoral prostheses. Methods. A retrospective radiological analysis was performed on adult patients with distal femoral arthroplasties. In all, 16 patients were included in the chemotherapy group with 18 patients in the non-chemotherapy control group. Annual radiographs were analyzed for three years postoperatively. Dimensions of the bony pedicle, osseointegration of the hydroxyapatite (HA) collar surface, bone resorption at the implant shoulder, and radiolucent line (RLL) formation around the cemented component were analyzed. Results. A greater RLL score (p = 0.041) was observed at three years postoperatively, with those receiving chemotherapy showing greater radiological loosening compared with those not receiving chemotherapy. Chemotherapy patients experience osteolysis at the shoulder of the ingrowth collar over time (p < 0.001) compared with non-chemotherapy patients where osteolysis was not observed. A greater median percentage integration of the collar surface was observed in the non-chemotherapy group (8.6%, interquartile range (IQR) 0.0% to 37.9%; p = 0.021) at three years. Bone growth around the collar was observed in both groups, and no statistical difference in amount of extracortical bony bridging was seen. Conclusion. Multidrug chemotherapy affects the osseointegration of ingrowth collars and accelerates signs of radiological loosening. This may increase the risk of aseptic loosening in patients with massive segmental implants used to treat bone cancer. Cite this article: Bone Joint Res 2020;9(7):333–340

Aims. The aim of this study was to present the long-term surgical outcomes, complications, implant survival, and causes of implant failure in patients treated with the modified Harrington procedure using antegrade large diameter pins. Patients and Methods. A cohort of 50 consecutive patients who underwent the modified Harrington procedure for periacetabular metastasis or haematological malignancy between January 1996 and April 2018 were studied. The median follow-up time for all survivors was 3.2 years (interquartile range 0.9 to 7.6 years). Results. The five-year overall survival rate was 33% for all the patients. However, implant survival rates were 100% and 46% at five and ten years, respectively. Eight patients survived beyond five years. There was no immediate perioperative mortality or complications. A total of 15 late complications occurred in 11 patients (22%). Five patients (10%) required further surgery to treat complications. The most frequent complication was pin breakage without evidence of acetabular loosening (6%). Two patients (4%) underwent revision for aseptic loosening at 6.5 and 8.9 years after surgery. Ambulatory status and pain level were improved in 83% and 89%, respectively. Conclusion. The modified Harrington procedure for acetabular destruction has low complication rates, good functional outcome, and improved pain relief in selected patients. Cite this article: Bone Joint J 2019;101-B:1557–1562

Aims. The purpose of this study was to report the long-term results of extendable endoprostheses of the humerus in children after the resection of a bone sarcoma. Methods. A total of 35 consecutive patients treated with extendable endoprosthetic replacement of the humerus in children were included. There were 17 boys and 18 girls in the series with a median age at the time of initial surgery of nine years (interquartile range (IQR) 7 to 11). Results. The median follow-up time was 10.6 years (IQR 3.9 to 20.4). The overall implant survival at ten years was 75%. Complications occurred in 13 patients (37%). Subluxation at the proximal humerus occurred in 19 patients (54%) but only six (17%) were symptomatic. Subluxation was seen more commonly in children under the age of nine years (86%) than in those aged nine years or more (33%) (p = 0.002). Implant failure occurred in nine patients (26%): the most common cause was aseptic loosening (four patients, 11%). Lengthening of the implant was carried out in 23 patients (66%). At final follow up, three patients had a limb that was shortened by 5 cm or more. The mean Musculoskeletal Tumor Society (MSTS) functional score was 79% (73% to 90%). Conclusion. Extendable endoprosthetic replacement is a reliable method of reconstructing humerus after excision of a bone sarcoma. Children who are less than nine years old have a high risk of subluxation. Cite this article: Bone Joint J 2020;102-B(1):64–71

Aims. The aim of this study was to report the results of custom-made endoprostheses with extracortical plates plus or minus a short, intramedullary stem aimed at preserving the physis after resection of bone sarcomas in children. Patients and Methods. Between 2007 and 2017, 18 children aged less than 16 years old who underwent resection of bone sarcomas, leaving ≤ 5 cm of bone from the physis, and reconstruction with a custom-made endoprosthesis were reviewed. Median follow-up was 67 months (interquartile range 45 to 91). The tumours were located in the femur in 11 patients, proximal humerus in six, and proximal tibia in one. Results. The five-year overall survival rate was 78%. No patient developed local recurrence. The five-year implant survival rate was 79%. In all, 11 patients (61%) developed a complication. Seven patients (39%) required further surgery to treat the complications. Implant failures occurred in three patients (17%) including one patient with aseptic loosening and two patients with implant or periprosthetic fracture. The preserved physis continued to grow at mean 3.3 cm (0 to 14). The mean Musculoskeletal Society score was 88% (67% to 97%). Conclusion. Custom-made endoprostheses that aim to preserve the physis are a safe and effective option for preserving physeal growth, limb length, and joint function with an acceptable rate of complications. Cite this article: Bone Joint J 2019;101-B:1144–1150

We have investigated whether improvements in design have altered the outcome for patients undergoing endoprosthetic replacement of the proximal tibia following resection of a tumour. Survival of the implant and ‘servicing’ procedures have been documented using a prospective database. A total of 194 patients underwent a proximal tibial replacement, with 95 having a fixed-hinge design and 99 a rotating-hinge with a hydroxyapatite collar; their median age was 21.5 years (10 to 74). At a mean follow-up of 14.7 years (5 to 29), 115 patients remain alive. The risk of revision for any reason in the fixed-hinge group was 32% at five years, 61% at ten years and 75% at 15 and 20 years, and in the rotating-hinge group 12% at five years, 25% at ten years and 30% at 15 years. Aseptic loosening was the most common reason for revision in the fixed-hinge knees, fracture of the implant in the early design of rotating hinges and infection in the current version. The risk of revision for aseptic loosening in the fixed-hinge knees was 46% at ten years. This was reduced to 3% in the rotating-hinge knee with a hydroxyapatite collar. The cemented, rotating hinge design currently offers the best chance of long-term survival of the prosthesis

We investigated whether improvements in design have altered the outcome for patients undergoing endoprosthetic replacement of the distal femur after resection of a tumour. Survival of the implant and ‘servicing’ procedures have been documented using a prospective database, review of the design of the implant and case records. In total, 335 patients underwent a distal femoral replacement, 162 having a fixed-hinge design and 173 a rotating-hinge. The median age of the patients was 24 years (interquartile range 17 to 48). A total of 192 patients remained alive with a mean follow-up of 12 years (5 to 30). The risk of revision for any reason was 17% at five years, 33% at ten years and 58% at 20 years. Aseptic loosening was the main reason for revision of the fixed-hinge knees while infection and fracture of the stem were the most common for the rotating-hinge implant. The risk of revision for aseptic loosening was 35% at ten years with the fixed-hinge knee, which has, however, been replaced by the rotating-hinge knee with a hydroxyapatite collar. The overall risk of revision for any reason fell by 52% when the rotating-hinge implant was used. Improvements in the design of distal femoral endoprostheses have significantly decreased the need for revision operations, but infection remains a serious problem. We believe that a cemented, rotating-hinge prosthesis with a hydroxyapatite collar offers the best chance of long-term survival of the prosthesis

We have made a retrospective study of 1001 custom-made prostheses used as replacements after surgery for bone tumours. There were 493 distal femoral, 263 proximal femoral and 245 proximal tibial prostheses. Aseptic loosening was shown to be the principal mode of failure of the implants, and 71 patients had revision for aseptic loosening of a cemented intramedullary stem. The probability of a patient surviving aseptic loosening for 120 months was 93.8% for a proximal femoral replacement, 67.4% for a distal femoral prosthesis and 58% for a proximal tibial implant. In patients with distal femoral replacements the age of the patient at the time of operation and the percentage of bone resected were related to the risk of aseptic loosening. Young patients with distal femoral prostheses in whom a high percentage of the femur had been replaced had the poorest prognosis for survival without aseptic loosening. The percentage of bone removed had a significant effect in the proximal tibial replacement group, but the age of the patient did not. By contrast, neither the age nor the percentage of bone removed was a factor after proximal femoral replacement. The significance of these findings is discussed in relation to mechanical factors

Aims. The aim of this study was to evaluate the prosthesis characteristics and associated conditions that may modify the survival of total femoral endoprosthetic replacements (TFEPR). Patients and Methods. In all, 81 patients treated with TFEPR from 1976 to 2017 were retrospectively evaluated and failures were categorized according to the Henderson classification. There were 38 female patients (47%) and 43 male patients (53%) with a mean age at diagnosis of 43 years (12 to 86). The mean follow-up time was 10.3 years (0 to 31.7). A survival analysis was performed followed by univariate and multivariate Cox regression to identify independent implant survival factors. Results. The revision-free survival of the implant was 71% at five years and 63.3% at ten years. Three prostheses reached 15 years without revision. The mean Musculoskeletal Tumor Society score in the group was 26 (23 to 28). The mechanisms of failure were infection in 18%, structural failures in 6%, tumour progression in 5%, aseptic loosening in 2%, and soft-tissue failures in 1%. Prostheses used for primary reconstruction after oncological resections had lower infection rates than revision implants (8% vs 25%; p = 0.001). The rates of infection in silver-coated and non-silver-coated prosthesis were similar (17.4% vs 19.%; p = 0.869). The incidence of hip dislocation was 10%. Rotating hinge prosthesis had a lower failure rate than fixed hinge prosthesis (5.3% vs 11%). After Cox regression, the independent factors associated with failures were the history of previous operations (hazard ratio (HR) 3.7; p = 0.041), and the associated arthroplasty of the proximal tibia (HR 3.8; p = 0.034). At last follow-up, 11 patients (13%) required amputation. Conclusion. TFEPR offers a reliable reconstruction option for massive bone loss of the femur, with a good survival when the prosthesis is used as a primary implant. The use of a rotating hinge at the knee and dual mobility bearing at the hip may be adequate to reduce the risk of mechanical and soft-tissue failures. Infection remains the main concern and there is insufficient evidence to support the routine use of silver-coated endoprosthesis. Cite this article: Bone Joint J 2019;101-B:522–528

In 251 patients over a period of 15 years an uncemented Kotz modular femoral and tibial reconstruction mega prosthesis was implanted after resection of a malignant tumour of the lower limb. Twenty-one patients (8.4%) underwent revision for aseptic loosening, again using an uncemented prosthesis, and five of these required a further revision procedure. The median follow-up time from the first revision was 60 months (11 to 168) and after a second revision, 33 months (2 to 50). The probability of a patient avoiding aseptic loosening for ten years was 96% for a proximal femoral, 76% for a distal femoral and 85% for a proximal tibial implant. At the time of follow-up all radiographs were assessed according to the International Symposium of Limb Salvage criteria. The first radiological signs of aseptic loosening were always seen at the most proximal or distal part of the anchorage stem at a mean of 12 months (4 to 23) after the first implantation. Using the Musculoskeletal Tumor Society score for evaluation, the clinical results showed a mean of 88% of normal function

We determined the efficacy of a devitalised autograft (n = 13) and allograft (n = 16) cortical strut bone graft combined with long-stem endoprosthetic reconstruction in the treatment of massive tumours of the lower limb. A total of 29 patients (18 men:11 women, mean age 20.1 years (12 to 45) with a ratio of length of resection to that of the whole prosthesis of > 50% were treated between May 2003 and May 2012. The mean follow-up was 47 months (15 to 132). The stem of the prosthesis was introduced through bone graft struts filled with cement, then cemented into the residual bone. Bone healing was achieved in 23 patients (86%). The mean Musculoskeletal Tumour Society functional score was 85% (57 to 97). The five-year survival rate of the endoprostheses was 81% (95% confidence intervals 67.3 to 92.3). The mean length of devitalised autografts and allografts was 8.6 cm (5 to 15), which increased the ratio of the the length of the stem of the prosthesis to that of the whole length of the prosthesis from a theoretical 35% to an actual 55%. . Cortical strut bone grafting and long-stem endoprosthetic reconstruction is an option for treating massive segmental defects following resection of a tumour in the lower limb. Patients can regain good function with a low incidence of aseptic loosening. The strut graft and the residual bone together serve as a satisfactory bony environment for a revision prosthesis, if required, once union is achieved. Cite this article: Bone Joint J 2015;97-B:544–9

Aims

The aim of this study was to investigate the safety and efficacy of 3D-printed modular prostheses in patients who underwent joint-sparing limb salvage surgery (JSLSS) for malignant femoral diaphyseal bone tumours.

Aims

Limb salvage surgery (LSS) is the primary treatment option for primary bone malignancy. It involves the removal of bone and tissue, followed by reconstruction with endoprosthetic replacements (EPRs) to prevent amputation. Trabecular metal (TM) collars have been developed to encourage bone ingrowth (osseointegration (OI)) into EPRs. The primary aim of this study was to assess whether OI occurs when TM collars are used in EPRs for tumour.

We evaluated the clinical results and complications after extra-articular resection of the distal femur and/or proximal tibia and reconstruction with a tumour endoprosthesis (MUTARS) in 59 patients (mean age 33 years (11 to 74)) with malignant bone or soft-tissue tumours. According to a Kaplan–Meier analysis, limb survival was 76% (95% confidence interval (CI) 64.1 to 88.5) after a mean follow-up of 4.7 years (one month to 17 years). Peri-prosthetic infection was the most common indication for subsequent amputation (eight patients). Survival of the prosthesis without revision was 48% (95% CI 34.8 to 62.0) at two years and 25% (95% CI 11.1 to 39.9) at five years post-operatively. Failure of the prosthesis was due to deep infection in 22 patients (37%), aseptic loosening in ten patients (17%), and peri-prosthetic fracture in six patients (10%). Wear of the bearings made a minor revision necessary in 12 patients (20%). The mean Musculoskeletal Tumor Society score was 23 (10 to 29). An extensor lag > 10° was noted in ten patients (17%). These results suggest that limb salvage after extra-articular resection with a tumour prosthesis can achieve good functional results in most patients, although the rates of complications and subsequent amputation are higher than in patients treated with intra-articular resection. Cite this article: Bone Joint J 2013;95-B:1425–31