To simulate the spinal instability that is found clinically after anterior corpectomy for the treatment of a fracture or a neoplasm, twelve fresh calf-spine segments, each containing five motion segments, were destabilized using a complete anterior corpectomy at the third lumbar level and anterior discectomies at the second and third and the third and fourth lumbar levels. Mechanical non-destructive cyclical testing in axial compression, rotation, and flexion was performed on each spinal segment after stabilization was accomplished. The three anterior-stabilization constructs that were compared were: (1) iliac strut grafting, (2) polymethylmethacrylate and anterior Harrington-rod instrumentation (the technique of Siegal and Siegal), and (3) the Kaneda anterior device. After anterior iliac-crest strut grafting, four types of posterior instrumentation were also tested sequentially: (1) Harrington distraction rods, (2) Luque rectangular instrumentation, (3) Cotrel-Dubousset transpedicular instrumentation, and (4) Steffee transpedicular screws and plates. Rotation, torque, axial displacement, and axial loads were measured during loading across the whole spinal segment between the grip points. Using an anterior extensometer, intervertebral displacement at the second, third, and fourth lumbar levels, and thus across the corpectomy defect at the third lumbar level, was recorded "on line" during testing in flexion and axial load. By recording the intervertebral displacement, the efficacy of each spinal construct in minimizing motion across the corpectomy defect could be quantified. The value for one-way analysis of variance for axial intervertebral displacement across the site of the third lumbar corpectomy was F = 10.5, p less than 0.001. The value for one-way analysis of variance for flexural intervertebral displacement across the corpectomy defect was F = 21.1, p less than 0.001. Homogeneous subsets of rigidity for torsional stiffness revealed that the least rigid constructs were iliac grafting alone, Harrington-rod instrumentation, and Luque rectangular instrumentation. The most rigid constructs were the anterior Kaneda device, transpedicular Cotrel-Dubousset instrumentation, and Steffee screws and plates. CLINICAL RELEVANCE: After corpectomy, spinal reconstructive surgery can restore axial, torsional, and flexural rigidity to normal levels. These experimental conclusions applied to the acute restoration of stability, rather than to rigidity after long-term cyclical loading. Using the most rigid anterior system, the Kaneda device, the fixation extended only one vertebral level cephalad and one level caudad to the corpectomy defect.(ABSTRACT TRUNCATED AT 400 WORDS)