Impacts of Tumor Location, Nature and Bone Destruction of Extremity Osteosarcoma on Selection of Limb Salvage Operative Procedure.

Objective: To study the impacts of tumor location, nature and extent of bone destruction on selection of operative protocol for extremity osteosarcoma (OS).

Methods: The medical records of 201 patients with extremity OS treated in our institute from December 1999 to June 2014 were retrospectively reviewed. Ninety eligible patients (56 males and 34 females) of average age 20 ± 11 years (range, 4-40 years) were enrolled. Tumor locations were categorized as diaphyseal (4; diaphysis group) or juxta-articular (86); the latter being subclassified as with (51, type III, epiphysis group) or without invasion beyond the epiphyseal line or plate (35, type I and II, metaphysis group) according to MRI images. Tumor nature (osteogenic, 51; osteolytic, 39) was determined radiologically. Extent of bone destruction was quantitated according to Mirel's scoring system to obtain an "invasion score". Regular postoperative follow-up included physical examination and imaging evaluation.

Results: Fifty-four patients underwent biological reconstruction and 36 mechanical reconstruction. The mean follow-up duration was 51 months (range, 6-176 months, including four deaths within 12 months). Biological reconstruction was performed more frequently in the diaphysis and metaphysis groups (31/39, 79.5%) than mechanical reconstruction (8/39, 20.5%, P < 0.05). Biological reconstruction and articular preservation were associated with more satisfactory limb function (MSTS scores: 25.0 ± 3.3 and 25.1 ± 3.6) than mechanical reconstruction and articular resection (MSTS scores: 23.4 ± 3.7 and 23.1 ± 3.4, P < 0.05). Reconstruction methods and articular preservation had no relationship with overall or tumor-free survival (P > 0.05). Osteolytic lesions were associated with more extensive bone destruction than osteogenic lesions according to invasion scores (P < 0.05). Following biological reconstruction, high invasion scores (>8) had a 13.5-fold risk of fracture compared with low scores (≤8) (P < 0.05). Twenty-one subjects had recurrences, 30 metastases and 26 died. Postoperative complications included infection (6), fracture (10), and prosthesis loosening (4). Kaplan-Meier analysis indicated 5- and 10-year survival rates of 68.9% and 62.8%, respectively, and 5- and 10-year tumor-free survival rates of 66.7% and 57.8%, respectively.

Conclusion: Selection of limb salvage operative protocol for extremity OS should rely on tumor location, nature and extent of bone destruction. Regardless of tumor site, mechanical reconstruction is indicated for tumors with high invasion scores (>8), whereas biological reconstruction is preferred for those with low invasion scores (≤8). Tumors sparing the epiphyseal line or plate are ideal candidates for articular preservation.