The impact of amount of bone resection on uncemented prosthesis failure in patients with a distal femoral tumor

Abstract

Background and objectives: Previous reports on the correlation between the amount of bone resection and prosthetic failure might be confounded by variability in the amount of soft tissue resected and by the mode of prosthesis fixation.

Methods: We analyzed 117 patients who underwent intra-articular resection and cementless modular tumor prosthetic reconstruction for a distal femoral tumor. Mean follow-up duration was 95 months (range, 15-271 months). Associations between the prognostic variables and prosthesis survival were assessed. A receiver operating characteristic (ROC) curve was plotted for resection percentage to predict prosthetic failure.

Results: Prostheses were removed in 35 (30%) patients for; infection (17), local recurrence (5), loosening (7), stem fracture (4), or periprosthetic fracture (2). The 10-year prosthetic survivals of the 117 implants were 65.2 ± 5.4%. A percentage bone resection (>40%) was found to be associated with mechanical prosthesis failure by multivariate analysis (P = 0.003). ROC curve analysis demonstrated that an optimal cut-off point of 43% for resection percentage had a sensitivity of 95.9% and a specificity of 45.5% for predicting mechanical prosthesis failure.

Conclusions: Infection is the major cause of prosthetic failure, and amount of bone resection is negatively correlated with implant longevity.