Study design: Single- or dual-shape memory metal (nitinol) rods were used to create spinal deformity in a mini-pig model.
Objective: To determine the practicality of employing nitinol rods for clinical spinal deformity correction using an animal deformity creation model.
Summary of background data: Nitinol rods can, theoretically, be approximated to curved spines when cooled/malleable, achieving spinal deformity correction on warming and returning to their original (straight) shape. Square cross-sectional rods may allow transverse plane correction. Single-rod constructs could achieve greater deformity correction clinically, whereas dual rods provide lower implant failure risk. The clinical feasibility and potential effectiveness of single versus dual nitinol rod constructs in creating spinal deformity in an animal model was studied.
Methods: Twenty mature mini-pigs were grouped: single rod with screws locked to the rod, single rod with screws unlocked, dual rods--locked, dual rods--unlocked. Square nitinol rods (80° manufactured curvature) were cooled (-20°C), straightened, placed into multilevel square-headed pedicle screws and warmed (40°C). Serial radiographs followed deformity creation over 12 weeks. Single versus dual rods, locked versus unlocked rods were compared (P<0.01). Computed tomography showed screw placement and fusion (none attempted). Preoperative and 12-week postoperative serum nickel levels were compared (P<0.05).
Results: Scoliotic deformities were created immediately in single and dual rod groups (28°±8°, 26°±7°; P=0.59); locked and unlocked groups (24°±7°, 30°±6°; P=0.08). At final follow-up, there was no difference between single and dual rods (31°±11°, 28°±10°; P=0.58); unlocked rods and locked rods were 34.9°±9.4° and 25.0°±8.1° (P=0.02). No implant failure occurred, however, the aggressive rod contour led to vertebral endplate fractures. Serum nickel 12 weeks post rod placement were unchanged from preop levels (5.1±0.6 μg/L, 4.7±0.2 μg/L, P=0.10).
Conclusion: Nitinol rods, reliable in creating spinal deformity in an animal model, could potentially straighten deformed spines. Square rods, facilitated by markedly reduced stiffness when cooled, have the potential to predictably perform axial plane correction.