Survival, complications and functional outcomes of cemented megaprostheses for high-grade osteosarcoma around the knee.

PURPOSE: We initiated a retrospective study on the long-term survival of cemented endoprostheses for bone tumours around the knee to answer the following questions: (1) What was the survival of these patients? (2) What was the overall survival of cemented prostheses around the knee? (3) What types of failures were observed in these reconstructions? (4) Did the survival and complications vary according to the site of the implant? (5) What was the functional result after cemented prosthesis replacement around the knee?

METHODS: From January 2006 to December 2013, 108 consecutive patients with an average age of 25 years, who had mature bone development as evidenced by imaging examinations, underwent 108 cemented endoprosthetic knee replacements for osteosarcoma resection. All patients received neoadjuvant chemotherapy using a multi-drug protocol consisting of high dose methotrexate (HDMTX), doxorubicin (ADM), cisplatin (DDP) and high dose ifosfamide (HDIFO). When extensor mechanism reconstruction was required, we ran nonabsorbable sutures through designated holes in the tibial component to fix detached hamstrings and the remaining ligaments in an imbricated fashion as well as reinforced the reconstruction with a medial gastrocnemius flap. Seventy-two (72/108, 66.7%) lesions were located in the distal femur and 36 (36/108, 33.3%) lesions at the proximal tibias. Nineteen patients were staged as IIA and 89 as IIB according to the Enneking staging system. The average follow-up was 53.3 months (range 12-125 months), with a minimum oncological follow-up of one year. Survival, prosthetic failure, complications and functional outcomes were recorded and reassessed at every visit after the primary operation.

RESULTS: At the final follow-up, the oncologic results showed that 33 patients died from metastases, and local recurrence occurred in ten patients. The estimated overall five-year and eight-year survival rates were 71% (95% CI: 62.4-79.65%) and 67.2% (95% CI: 58-76.4%), respectively. In this study, a total of 51 complications occurred in 45 patients, and at the end of follow-up, 59 patients had prostheses in situ. The estimated overall five-year and eight-year implant survival rates were 77.7% (95% CI: 67.9-87.5%) and 54.5% (95% CI: 31.4-77.6%), respectively, when patients who died with their original prostheses were censored. In total, 21 (21/108, 19.4%) implants failed, five due to infections (5/21), eight due to aseptic loosening (8/21), four due to local recurrence (4/21), three due to structural failure (3/21) and one due to soft tissue failure (1/21). The average Musculoskeletal Tumor Society Score (MSTS) at the most recent follow-up was 22.9 (9-30) points on a 30-point scale, which indicated an excellent or good functional outcome. Analysis of the results based on implant site revealed a slight difference of the estimated five-year prosthesis survival between implants located in the distal femur and those located in the proximal tibia [86.1% (95% CI: 75.5-97%) versus 66.9% (95% CI: 49.8-83.9%); P = 0.09]. However, the functional outcomes and complication rates of prostheses located in the distal femur were both better than those located in the proximal tibia.

CONCLUSIONS: With effective management strategies for complications, cemented endoprosthetic reconstruction of the proximal tibia and distal femur using an extensor mechanism reconstruction technique provides a reliable method of reconstruction following tumour resection around the knee. Level of Evidence Level IV, therapeutic study.