Aims. Venous tumour thrombus (VTT) is a rare finding in osteosarcoma. Despite the high rate of VTT in osteosarcoma of the pelvis, there are very few descriptions of VTT associated with extrapelvic primary osteosarcoma. We therefore sought to describe the prevalence and presenting features of VTT in osteosarcoma of both the pelvis and the limbs. Methods. Records from a single institution were retrospectively reviewed for 308 patients with osteosarcoma of the pelvis or limb treated between January 2000 and December 2022. Primary lesions were located in an upper limb (n = 40), lower limb (n = 198), or pelvis (n = 70). Preoperative imaging and operative reports were reviewed to identify patients with thrombi in proximity to their primary lesion. Imaging and histopathology were used to determine presence of tumour within the thrombus. Results. Tumours abutted the blood vessels in 131 patients (43%) and encased the vessels in 30 (10%). Any form of venous thrombus was identified in 31 patients (10%). Overall, 21 of these thrombi were determined to be involved with the tumour based on imaging (n = 9) or histopathology (n = 12). The rate of VTT was 25% for pelvic osteosarcoma and 1.7% for limb osteosarcoma. The most common imaging features associated with histopathologically proven VTT were enhancement with contrast (n = 12; 100%), venous enlargement (n = 10; 83%), vessel encasement (n = 8; 66%), and visible intraluminal osteoid matrix (n = 6; 50%). Disease-specific survival (DSS) for patients with VTT was 95% at 12 months (95% CI 0.87 to 1.00), 50% at three years (95% CI 0.31 to 0.80), and 31% at five years (95% CI 0.14 to 0.71). VTT was associated with worse DSS (hazard ratio 2.3 (95% CI 1.11 to 4.84). Conclusion. VTT is rare with osteosarcoma and occurs more commonly in the pelvis than the limbs. Imaging features suggestive of VTT include enhancement with contrast, venous dilation, and vessel encasement. VTT portends a worse prognosis for patients with osteosarcoma, with a similar survivability to metastatic disease. Cite this article: Bone Joint J 2024;106-B(8):865–870

Aims. Low-grade central osteosarcoma (LGCOS), a rare type of osteosarcoma, often has misleading radiological and pathological features that overlap with those of other bone tumours, thereby complicating diagnosis and treatment. We aimed to analyze the clinical, radiological, and pathological features of patients with LGCOS, with a focus on diagnosis, treatment, and outcomes. Methods. We retrospectively analyzed the medical records of 49 patients with LGCOS (Broder’s grade 1 to 2) treated between January 1985 and December 2017 in a single institute. We examined the presence of malignant features on imaging (periosteal reaction, cortical destruction, soft-tissue invasion), the diagnostic accuracy of biopsy, surgical treatment, and oncological outcome. Results. Based on imaging, 35 of 49 patients (71.4%) exhibited malignant features. Overall, 40 of 49 patients (81.6%) had undergone a biopsy before en-bloc resection: 27 of 40 patients (67.5%) were diagnosed on the first biopsy, which was more accurate when carried out by open rather than needle biopsy (91.3% vs 35.3% diagnostic accuracy, respectively; p < 0.001). Of the 40 patients treated by en-bloc resection, surgical margins were wide in 38 (95.0%) and marginal in two (5.0%). Furthermore, nine of 49 patients (18.4%) underwent curettage (intralesional margin) without previous biopsy. All patients with a positive margin developed local recurrence. Distant metastases occurred in five of 49 patients (10.2%). The mean five-year overall survival (OS) and distant relapse-free survival (D-RFS) were 89.3% (SD 5.1%) and 85.7% (SD 5.5%), respectively. Univariate analysis showed that the occurrence of distant metastasis was a poor prognostic factor for OS (hazard ratio 11.54, 95% confidence interval (CI) 1.92 to 69.17; p < 0.001). Local recurrence was a poor prognostic factor for D-RFS (HR 8.72, 95% CI 1.69 to 45.0; p = 0.002). Conclusion. The diagnosis of LGCOS can be challenging because it may present with non-malignant features and has a low diagnostic accuracy on biopsy. If precisely diagnosed, LGCOS can be successfully treated by surgical excision with wide margins. Cite this article: Bone Joint J 2024;106-B(1):99–106

Aims. We have previously reported cryoablation-assisted joint-sparing surgery for osteosarcoma with epiphyseal involvement. However, it is not clear whether this is a comparable alternative to conventional joint arthroplasty in terms of oncological and functional outcomes. Methods. A total of 22 patients who had localized osteosarcoma with epiphyseal involvement around the knee and underwent limb salvage surgery were allocated to joint preservation (JP) group and joint arthroplasty (JA) group. Subjects were followed with radiographs, Musculoskeletal Tumor Society (MSTS) score, and clinical evaluations at one, three, and five years postoperatively. Results. Patients in both groups (ten in JP and 12 in JA) did not differ in local recurrence (p ≥ 0.999) and occurrence of metastases (p ≥ 0.999). Overall survival was similar in both groups (p = 0.858). Patients in the JP group had less range of motion (ROM) of the knee (p < 0.001) and lower MSTS scores (p = 0.010) compared with those of the JA group only at one year postoperatively. There was no difference between groups either at three years for ROM (p = 0.185) and MSTS score (p = 0.678) or at five years for ROM (p = 0.687) and MSTS score (p = 0.536), postoperatively. Patients in the JA group tended to have more complications (p = 0.074). Survival of primary reconstruction in the JP group was better than that of the JA group (p = 0.030). Conclusion. Cryoablation-aided joint-sparing surgery offers native joint preservation with comparable functional recovery and more durable reconstruction without jeopardizing oncological outcomes compared with conventional limb salvage surgery. Cite this article: Bone Joint J 2021;103-B(8):1421–1427

Aims . Osteosarcoma of the pelvis is a particularly difficult tumour to treat as it often presents late, may be of considerable size and/or associated with metastases when it presents, and is frequently chondroid in origin and resistant to chemotherapy. The aim of this study was to review our experience of managing this group of patients and to identify features predictive of a poor outcome. Patients and Methods. Between 1983 and 2014, 121 patients, (74 females and 47 males) were treated at a single hospital: 74 (61.2%) patients had a primary osteosarcoma and 47 (38.8%) had an osteosarcoma which was secondary either to Paget’s disease (22; 18.2%) or to previous pelvic irradiation (25; 20.7%). . The mean age of those with a primary osteosarcoma was 29.3 years (nine to 76) and their mean follow-up 2.9 years (0 to 29). The mean age of those with a secondary sarcoma was 61.9 years (15 to 85) and their mean follow-up was one year (0 to 14). . A total of 22 patients with a primary sarcoma (52.4%) and 20 of those with a secondary sarcoma (47.6%) had metastases at the time of presentation. . Results. The disease-specific survival at five years for all patients was 27.2%. For those without metastases at the time of diagnosis, the five-year survival was 32.7%. Factors associated with a poor outcome were metastases at diagnosis and secondary tumours. In primary osteosarcoma, sacral location, surgical margin and a diameter > 10 cm were associated with a poor outcome. Conclusion. In this, the largest single series of patients with an osteosarcoma of the pelvis treated in a single hospital, those with secondary tumours and those with metastases at presentation had a particularly poor outcome. For those with a primary sarcoma, sacral location, an intralesional margin and a diameter of > 10 cm were poor prognostic indicators. Cite this article: Bone Joint J 2016;98-B:555–63

Aims. The aim of this study was to determine the risk of local recurrence and survival in patients with osteosarcoma based on the proximity of the tumour to the major vessels. Patients and Methods. A total of 226 patients with high-grade non-metastatic osteosarcoma in the limbs were investigated. Median age at diagnosis was 15 years (4 to 67) with the ratio of male to female patients being 1.5:1. The most common site of the tumour was the femur (n = 103) followed by tibia (n = 66). The vascular proximity was categorized based on the preoperative MRI after neoadjuvant chemotherapy into four types: type 1 > 5 mm; type 2 ≤ 5 mm, > 0 mm; type 3 attached; type 4 surrounded. Results. Limb salvage rate based on the proximity type was 92%, 88%, 51%, and 0% for types 1 to 4, respectively, and the overall survival at five years was 82%, 77%, 57%, and 67%, respectively (p < 0.001). Local recurrence rate in patients with limb-salvage surgery was 7%, 8%, and 22% for the types 1 to 3, respectively (p = 0.041), and local recurrence at the perivascular area was observed in 1% and 4% for type 2 and 3, respectively. The mean microscopic margin to the major vessels was 6.9 mm, 3.0 mm, and 1.4 mm for types 1 to 3, respectively. In type 3, local recurrence-free survival with limb salvage was significantly poorer compared with amputation (p = 0.025), while the latter offered no overall survival benefit. In this group of patients, factors such as good response to chemotherapy or limited vascular attachment to less than half circumference or longitudinal 10 mm reduced the risk of local recurrence. Conclusion. The proximity of osteosarcoma to major blood vessels is a poor prognostic factor for local control and survival. Amputation offers better local control for tumours attached to the blood vessels but does not improve survival. Limb salvage surgery offers similar local control if the tumour attachment to blood vessels is limited. Cite this article: Bone Joint J 2019;101-B:1024–1031

Aims. For paediatric and adolescent patients with growth potential, preservation of the physiological joint by transepiphyseal resection (TER) of the femur confers definite advantages over arthroplasty procedures. We hypothesized that the extent of the tumour and changes in its extent after neoadjuvant chemotherapy are essential factors in the selection of this procedure, and can be assessed with MRI. The oncological and functional outcomes of the procedure were reviewed to confirm its safety and efficacy. Methods. We retrospectively reviewed 16 patients (seven male and nine female, mean age 12.2 years (7 to 16)) with osteosarcoma of the knee who had been treated by TER. We evaluated the MRI scans before and after neoadjuvant chemotherapy for all patients to assess the extent of the disease and the response to treatment. Results. The mean follow-up period was 64.3 months (25 to 148) after surgery and no patients were lost to follow-up. On MRI evaluation, 13 tumours were near but not in contact with the physes and three tumours were partially in contact with the physes before neoadjuvant chemotherapy. Bone oedema in the epiphysis was observed in eight patients. After neoadjuvant chemotherapy, bone oedema in the epiphysis disappeared in all patients. In total, 11 tumours were not in contact and five tumours were in partial contact with the physes. The postoperative pathological margin was negative in all patients. At the last follow-up, 12 patients were continuously disease-free and three had no evidence of disease. One patient died due to the disease. Functionally, the patients with retained allograft or recycled autograft had a mean knee range of flexion of 126° (90° to 150°). The mean Musculoskeletal Tumor Society functional score was 27.6 (23 to 30). Conclusion. TER is an effective limb-salvage technique for treating malignant metaphyseal bone tumours in paediatric and young osteosarcoma patients with open physes when a good response to chemotherapy and no progression of the tumour to the epiphysis have been confirmed by MRI. Cite this article: Bone Joint J 2020;102-B(6):772–778

Aims. To assess the correlation between the histological response to preoperative chemotherapy and event-free survival (EFS) or overall survival (OS) in patients with high-grade localized osteosarcoma. Methods. Out of 625 patients aged ≤ 40 years treated for primary high-grade osteosarcoma between 1997 and 2016, 232 patients without clinically detectable metastases at the time of diagnosis and treated with preoperative high-dose methotrexate, adriamycin and cisplatin (MAP) chemotherapy and surgery were included. Associations of chemotherapy-induced necrosis in the resected specimen and EFS or OS were assessed using Cox model and the Pearson’s correlation coefficients (r). Time-dependent receiver operating characteristic analysis was applied to determine the optimal cut-off value of chemotherapy-induced necrosis for EFS and OS. Results. OS was 74% (95% confidence interval (CI) 67 to 79) at five years. Median chemotherapy-induced necrosis was 85% (interquartile range (IQR) 50% to 97%). In multivariate Cox model, chemotherapy-induced necrosis was significantly associated with EFS and OS (hazard ratio (HR) = 0.99 (95% CI 0.98 to 0.99); p < 0.001 and HR = 0.98 (95% CI 0.97 to 0.99); p < 0.001, respectively). Positive correlation was observed between chemotherapy-induced necrosis and five-year EFS and five-year OS (r = 0.91; p < 0.001, and r = 0.85; p < 0.001, respectively). The optimal cut-off value of chemotherapy-induced necrosis for five-year EFS and five-year OS was 85% and 72%, respectively. Conclusion. Chemotherapy-induced necrosis in the resected specimen showed positive correlation with EFS and OS in patients with high-grade localized osteosarcoma after MAP chemotherapy. In our analysis, optimal cut-off values of MAP chemotherapy-induced necrosis in EFS and OS were lower than the commonly used 90%, suggesting the need for re-evaluation of the optimal cut-off value through larger, international collaborative research. Cite this article: Bone Joint J 2020;102-B(6):795–803

Opinion remains divided as to whether the development of pathological fracture affects the prognosis of patients with an osteosarcoma of the extremities. We conducted a comprehensive systematic review and meta-analysis of papers which reported the outcomes of osteosarcoma patients with and without a pathological fracture. There were eight eligible papers for final analysis which reported on 1713 patients, of whom 303 (17.7%) had a pathological fracture. The mean age for 1464 patients in six studies was 23.2 years old (2 to 82). The mean follow-up for 1481 patients in seven studies was 90.1 months (6 to 240). The pooled estimates of local recurrence rates in osteosarcoma patients with and without pathological fractures were 14.4% (8.7 to 20.0) versus 11.4% (8.0 to 14.8). The pooled estimate of relative risk was 1.39 (0.89 to 2.20). The pooled estimates of five-year event-free survival rates in osteosarcoma patients with and without a pathological fracture were 49.3% (95% CI 43.6 to 54.9) versus 66.8% (95% CI 60.7 to 72.8). The pooled estimate of relative risk was 1.33 (1.12 to 1.59). There was no significant difference in the rate of local recurrence between patients who were treated by amputation or limb salvage. The development of a pathological fracture is a negative prognostic indicator in osteosarcoma and is associated with a reduced five-year event-free survival and a possibly higher rate of local recurrence. Our findings suggest that there is no absolute indication for amputation, as similar rates of local recurrence can be achieved in patients who are carefully selected for limb salvage. . Cite this article: Bone Joint J 2014; 96-B:1396–1403

Aims. The use of frozen tumour-bearing autograft combined with a vascularized fibular graft (VFG) represents a new technique for biological reconstruction of massive bone defect. We have compared the clinical outcomes between this technique and Capanna reconstruction. Methods. From June 2011 to January 2016 a retrospective study was carried out of patients with primary osteosarcoma of lower limbs who underwent combined biological intercalary reconstruction. Patients were categorized into two groups based on the reconstructive technique: frozen tumour-bearing autograft combined with concurrent VFG (Group 1) and the Capanna method (Group 2). Demographics, operating procedures, oncological outcomes, graft union, limb function, and postoperative complications were compared. Results. A total of 23 patients were identified for analysis: eight in Group 1 and 15 in Group 2. There was no difference in the demographics (age, sex, and affected site) and operating procedures (resection length, duration of surgery, and blood loss) between the two groups. No significant difference was found in local recurrence in Group 1 versus Group 2 (p = 0.585). Mean union time for the frozen autograft-host junction was 8.4 months (7.0 to 11.0), significantly earlier than for the allograft-host junction in Group 2 (mean 14.1 months (10.0 to 28.0); p < 0.001). Mean Musculoskeletal Tumor Society scores in groups 1 and 2 were 90.3% (SD 7.4%) and 88.0% (SD 9.0%), respectively, with no significant statistical difference (p = 0.535). In terms of complications, infection (n = 1, 6.7%) and delayed union (n = 2, 13.3%) occurred in Group 2, but no such complications were observed in Group 1. Conclusion. Frozen tumour-bearing autograft in combination with VFG can be used as an alternative to the Capanna reconstruction in properly selected patients with osteosarcoma. Cite this article: Bone Joint J 2020;102-B(5):646–652

Aims. The aim of this study was to identify factors that determine outcomes of treatment for patients with chondroblastic osteosarcomas (COS) of the limbs and pelvis. Patients and Methods. The authors carried out a retrospective review of prospectively collected data from 256 patients diagnosed between 1979 and 2015. Of the 256 patients diagnosed with COS of the pelvis and the limbs, 147 patients (57%) were male and 109 patients (43%) were female. The mean age at presentation was 20 years (0 to 90). Results. In all, 82% of the patients had a poor response to chemotherapy, which was associated with the presence of a predominantly chondroblastic component (more than 50% of tumour volume). The incidence of local recurrence was 15%. Synchronous or metachronous metastasis was diagnosed in 60% of patients. Overall survival was 51% and 42% after five and ten years, respectively. Limb localization and wide surgical margins were associated with a lower risk of local recurrence after multivariable analysis, while the response to chemotherapy was not. Local recurrence, advanced patient age, pelvic tumours, and large volume negatively influenced survival. Resection of pulmonary metastases was associated with a survival benefit in the limited number of patients in whom this was undertaken. Conclusion. COS demonstrates a poor response to chemotherapy and a high incidence of metastases. Wide resection is associated with improved local control and overall survival, while excision of pulmonary metastases is associated with improved survival in selected patients. Cite this article: Bone Joint J 2019;101-B:739–744

Aims. The purpose of this study was to clarify the clinical behaviour, prognosis, and optimum treatment of dedifferentiated low-grade osteosarcoma (DLOS) diagnosed based on molecular pathology. Patients and Methods. We retrospectively reviewed 13 DLOS patients (six men, seven women; median age 32 years (interquartile range (IQR) 27 to 38)) diagnosed using the following criteria: the histological coexistence of low-grade and high-grade osteosarcoma components in the lesion, and positive immunohistochemistry of mouse double minute 2 homolog (MDM2) and cyclin-dependent kinase 4 (CDK4) associated with MDM2 amplification. These patients were then compared with 51 age-matched consecutive conventional osteosarcoma (COS) patients (33 men, 18 women; median age 25 years (IQR 20 to 38)) regarding their clinicopathological features. Results. The five-year overall survival (OAS) rates in the DLOS and COS patients were 85.7% and 77.1% (p = 0.728), respectively, and the five-year progression-free survival (PFS) rates were 57.7% and 44.9% (p = 0.368), respectively. A total of 12 DLOS patients received chemotherapy largely according to regimens for COS. Among the nine cases with a histological evaluation after chemotherapy, eight showed a poor response, and seven of these had a necrosis rate of < 50%. One DLOS patient developed local recurrence and five developed distant metastases. Conclusion. Based on our study of 13 DLOS cases that were strictly defined by histological and molecular means, DLOS showed a poorer response to a standard chemotherapy regimen than COS, while the clinical outcomes were not markedly different. Cite this article: Bone Joint J 2019;101-B:745–752

There are eight reported cases in the literature of osteosarcomas secreting β-hCG. Our primary aim was to investigate the rate of β-hCG expression in osteosarcoma and attempt to understand the characteristics of osteosarcomas that secrete β-hCG. We reviewed 37 histopathology slides (14 biopsies and 23 surgical specimens) from 32 patients with osteosarcoma. The slides were retrospectively stained for β-hCG expression. Patient and tumour characteristics, including age, gender, tumour location, subtype, proportion of necrosis, presence of metastases and recurrence were recorded. A total of five of the 32 tumours were found to be positive for β-hCG expression (one strongly and four weakly). This incidence of this expression was found in tumours with poor histological response to neoadjuvant chemotherapy. The use of β-hCG expression as a diagnostic, prognostic or follow-up marker is questionable and needs further investigation with a larger sample size

The aim of this study was to evaluate the prognostic and therapeutic factors which influence the oncological outcome of parosteal osteosarcoma. A total of 80 patients with a primary parosteal osteosarcoma were included in this retrospective study. There were 51 females and 29 males with a mean age of 29.9 years (11 to 78). The mean follow-up was 11.2 years (1 to 40). Overall survival was 91.8% at five years and 87.8% at ten years. Local recurrence occurred in 14 (17.5%) patients and was associated with intralesional surgery and a large volume of tumour. On histological examination, 80% of the local recurrences were dedifferentiated high-grade tumours. A total of 12 (14.8%) patients developed pulmonary metastases, of whom half had either a dedifferentiated tumour or a local recurrence. Female gender and young age were good prognostic factors. Local recurrence was a poor prognostic factor for survival. Medullary involvement or the use of chemotherapy had no impact on survival. The main goal in treating a parosteal osteosarcoma must be to achieve a wide surgical margin, as inadequate margins are associated with local recurrence. Local recurrence has a significant negative effect on survival, as 80% of the local recurrences are high-grade dedifferentiated tumours, and half of these patients develop metastases. The role of chemotherapy in the treatment of parosteal osteosarcoma is not as obvious as it is in the treatment of conventional osteosarcoma. The mainstay of treatment is wide local excision. Cite this article: Bone Joint J 2015;97-B:1698–1703

Between 1986 and 2002, 42 patients with synchronous multifocal osteosarcoma were treated with two different protocols of neoadjuvant chemotherapy. When feasible, the primary and secondary tumours were excised as a combined procedure. After initial chemotherapy 26 patients were excluded from simultaneous excision of all their secondary bone lesions as their disease was too advanced. In 12 patients only isolated excision of the primary lesion was possible. For 16 patients simultaneous operations were conducted to excise the primary and secondary lesions. This involved two supplementary sites in 15 patients and four additional sites in one patient. Of these, 15 attained remission but 12 relapsed and died (11 within two years). Three patients remained disease-free at five, six and 17 years. The histological response to pre-operative chemotherapy of the primary and secondary lesions was concordant in 13 of the 16 patients who underwent simultaneous operations at more than one site. The prognosis for synchronous multifocal osteosarcoma remains poor despite combined chemotherapy and surgery. The homogeneous histological responses in a large proportion of the primary and secondary lesions implies that synchronous multifocal osteosarcoma tumours are not multicentric in origin, but probably represent bone-to-bone metastases from a single tumour

We have reviewed the data from our regional Bone Tumour Registry on patients with osteosarcoma diagnosed between 1933 and 2004 in order to investigate the relationship between survival and changes in treatment. There were 184 patients with non-metastatic appendicular osteosarcoma diagnosed at the age of 18 or under. Survival was calculated using Kaplan-Meier curves, and multivariate analysis was performed using the Cox regression proportional hazards model. The five-year survival improved from 21% between 1933 and 1959, to 62% between 1990 and 1999. During this time, a multi-disciplinary organisation was gradually developed to manage treatment. The most significant variable affecting outcome was the date of diagnosis, with trends in improved survival mirroring the introduction of increasingly effective chemotherapy. Our experience suggests that the guidelines of the National Institute for Clinical Excellence on the minimum throughput of centres for treatment should be enforced flexibly in those that can demonstrate that their historical and contemporary results are comparable to those published nationally and internationally

We undertook a prospective study to evaluate the prognostic significance of the serum levels of vascular endothelial growth factor (VEGF) in predicting the survival of patients with osteosarcoma. The levels were measured by an enzyme-linked immunosorbent assay in 15 patients with osteosarcoma before commencing treatment. The patients were divided into two groups, with a high or a low serum VEGF level, and the incidence of metastases and overall survival rate were compared. No significant relationship was observed between the serum VEGF levels and gender, age, the size of the tumour or the response to pre-operative chemotherapy. Patients with a serum VEGF > 1000 pg/ml had significantly worse survival than those with a level < 1000 pg/ml (p = 0.002). The serum VEGF level may be useful in predicting the prognosis for survival in patients with osteosarcoma

We evaluated the long-term outcome of patients with an osteosarcoma who had undergone prior manipulative therapy, a popular treatment in Asia, and investigated its effects on several prognostic factors. Of the 134 patients in this study, 70 (52%) patients had manipulative therapy and 64 (48%) did not. The age, location, and size of tumour were not significantly different between the groups. The five-year overall survival rate was 58% and 92% in the groups with and without manipulative therapy (p = 0.004). Both the primary and overall rates of lung metastasis were significantly higher in the manipulative group (primary: 32% vs 3%, p = 0.003; overall lung metastasis rate: 51.4% vs 18.8%, p < 0.001). Patients who had manipulative therapy had higher local recurrence rates in comparison to patients who did not (29% vs 6%, p = 0.011). The prognosis for patients with osteosarcoma who had manipulative therapy was significantly poorer than those who had not. Manipulative therapy was an independent factor for survival. This form of therapy may serve as a mechanism to accelerate the spread of tumour cells, and therefore must be avoided in order to improve the outcome for patients with an osteosarcoma

We describe a case of osteosarcoma of the scaphoid bone, which to our knowledge is only the second reported case of osteosarcoma in the carpus. A 38-year-old man complained of intense pain in the right wrist and had curettage and a bone graft for a lesion in the scaphoid. Histological examination showed this to be an osteosarcoma. Below-elbow amputation was performed and adjuvant chemotherapy given. There has been no evidence of recurrence or metastases at 33 months after amputation

Aims. Children treated for osteosarcoma around the knee often have a substantial leg-length discrepancy at skeletal maturity. The aim of this study was to investigate the results of staged skeletal reconstruction after a leg lengthening procedure using an external fixator in these patients. Patients and Methods. We reviewed 11 patients who underwent staged reconstruction with either an arthroplasty (n = 6) or an arthrodesis (n = 5). A control group of 11 patients who had undergone wide excision and concurrent reconstruction with an arthroplasty were matched for gender, location, and size of tumour. We investigated the change in leg-length discrepancy, function as assessed by the Musculoskeletal Tumor Society Scale (MSTS) score and complications. Results. A mean 5.2 cm (1.7 to 8.9) of lengthening was achieved. The mean MSTS scores significantly improved after staged reconstruction (p = 0.003) but were still worse than those of the control group (p = 0.049). However, the MSTS scores of the arthroplasty subgroup were comparable with those of the controls, although the extensor lag was greater and the range of movement was less. The patient group experienced more complications, but all of these resolved. Conclusion. Approximately 5 cm of lengthening and significant functional improvement can be achieved by staged reconstruction and lengthening, without major complications. Although it has limitations, this method of treatment seems to be a satisfactory surgical option for growing children with a significant leg-length discrepancy after excision of an osteosarcoma around the knee. Cite this article: Bone Joint J 2017;99-B:401–8

We have studied 560 patients with osteosarcoma of a limb, who had been treated by neoadjuvant chemotherapy, in order to analyse the incidence of local and systemic recurrence according to the type of surgery undertaken. Of these, 465 patients had a limb-salvage procedure and 95 amputation or rotationplasty. At a median follow-up of 10.5 years there had been 225 recurrences. The five-year disease-free survival and overall survival rates were 60.7% and 68.5%, respectively, with no significant difference between patients undergoing amputation and those undergoing resection. The incidence of local recurrence was the same for patients treated by either amputation or limb salvage and correlated significantly with the margins of surgical excision and the histological response to chemotherapy. The outcome for patients with a local recurrence was significantly worse than for those who had recurrent disease with metastases only. We conclude that limb-salvage procedures are relatively safe in osteosarcoma treated by neoadjuvant chemotherapy. They should, however, only be performed in institutions where the margins of surgical excision and the histological response to chemotherapy can be accurately assessed. If the margins are inadequate and the histological response to chemotherapy is poor an immediate amputation should be considered