Long-term survival of allografts for reconstruction after resection of bone tumours of the proximal humerus T.J. Grundy, G. Squire, N.A. Ferran, W.M. Harper, R.U. Ashford Introduction Methods Results The proximal humerus is the third most common site for osteosarcoma, and the second most common site for bone metastasis in the appendicular skeleton. 1,2 Reconstruction is typically achieved by the use of an implant following resection. Implant options include autograft allograft, endoprosthetic replacement (EPR), and allograft–prosthetic composite (APC). 3 Allograft reconstruction is complicated by fracture, infection, and instability, frequently resulting in the need for implant revision or removal. 1,4 Our unit published early favourable results concerning allograft use, with all implants intact at a median follow-up of 23 months (14–112 months). 1 These results supported the use of allograft as an alternative to EPR and APC. We now present the long-term follow-up of patients who underwent tumour resection with massive humeral reconstruction using allograft implantation. Surgical outcome and complications In all cases surgical resection was complete and while there were no immediate complications post-operatively, patients have displayed extensive complications at follow-up. The two patients with metastatic disease died with their allograft intact, while all implants in patients with primary bone tumours were revised. Overall revision rate was 75%. Reasons for revision included non-union, fracture, infection, dislocation and local recurrence. (Figure 1) Survival The cumulative 10 year survival for patients was 60% and for implants was 0%. (Figure 2) This was a retrospective case review of patients undergoing massive proximal humeral allograft for primary and secondary bone tumours between 1991 and 2003. Patients were identified from the records of the Leicester bone bank. Case notes, histology records, clinical letters, and radiology reports were reviewed. Demographic data, histological diagnosis, relevant dates and reasons for revision were recorded in order to calculate patient and implant survival. Patients The median age at first surgery was 38.5 years (15–77 years), the male to female ration was 1:1, and median follow-up was 9.7 years (2–19.5 years). Six patients were treated for primary bone tumours and two for isolated metastases. Patients’ demographics are presented in Table 1. Oncological outcome In all cases surgical resection was adequate. The patient with metastatic RCC died 23 months post-operatively, and the patient with metastatic TC died 79 months post-operatively. Two primary tumours were found to recur, and a forequarter amputation was performed. Oncological outcome is presented in Table 1. Table 1. Demographic and oncological patient data. Pt Age Sex Diagnosis Oncological outcome Length of follow-up (years) 1 57 M RCC DOD 1.94 2 23 F OS ADF 19.6 3 32 F OS ADF 14.6 4 15 F OS DOD 11.6 5 77 M CS DOD 11.2 6 16 F OS DOD 8.1 7 45 M CS ADF 8.1 8 61 M TC DOD 6.6 (RCC – Renal Cell Carcinoma, OS – Osteosarcoma, CS – Chondrosarcoma, TC – Thyroid Carcinoma, DOD – Died of Disease, ADF – Alive and Disease Free) Figure 1. Radiographs demonstrating modes of failure; A: non-union, allograft resorbtion, and fracture, B: non-union secondary to infection, C: massive infection with metalwork failure. Figure 2. Kaplan-Meier survival curves for patients and implants. Implant vs Patient Survival Survival time (years) 0.0 5.0 10.0 15.0 20.0 0.0 0.2 0.4 0.6 0.8 1.0 Cum Survival Patient Implant Patient-censored Implant-censored Conclusions The goal of reconstructive tumour surgery is to adequately resect malignancy while restoring function for the lifetime of the patient. In this series 75% of implants failed within the lifetime of patients, thus failing to meet this goal. Proximal humerus allograft reconstruction is associated with a higher incidence of complications and failure as compared to APC, and EPR. Consequently we no longer recommend proximal humerus allograft reconstruction. References 1. R. U. Ashford et al. The Versatility of Massive Allografts in the Treatment of Bone Tumours of Proximal Humerus. European Journal of Orthopaedic Surgery & Traumatology 2004, 14: 234–238. 2. D. Dahlin. Bone Tumours: General Aspects and Data on 6,221 cases. Springfield (IL): Charles C. Thomas; 1978. 3. A. Black et al. Treatment of Malignant Tumours of the Proximal Humerus with Allograft-Prosthesis Composite reconstruction, Journal of Shoulder and Elbow Surgery, 2007, 16:525–533. 4. L. J. Probyn et al. A Comparison of Outcome of Osteoarticular Allograft Reconstruction and Shoulder Arthrodesis Following Resection of Primary Tumours of the Proximal Humerus, Sarcoma 1998, 2, 163–170.