Lumbar pedicle screw salvage: pullout testing of three different pedicle screw designs

Although research has determined pedicle screw pullout strengths for normal and osteoporotic bone, this study provides the first biomechanical analysis of pedicle screw salvage. Ten fresh frozen human lumbar spines were separated into individual vertebrae; 6.0 x 40 mm pedicle screws were placed in each pedicle; and an axial pullout test was performed to establish control values. Ultimate load, initial stiffness, work, and displacement data were calculated. Each vertebra was reinstrumented with one 7.0 x 40 mm variable screw placement (VSP) screw side by side with either a 7.0 mm Cotrel Dubousset sacral screw (CD) or a 7.0 mm Compact Cotrel Dubousset pedicle screw (CCD). Pullout tests were repeated and compared to control data for individual screws and for each VSP/CD or VSP/CCD pair. Vertebrae were then reinstrumented with 8.0 mm VSP and CD screws and paired pullouts repeated. Statistical analysis was carried out using a paired T test. Analysis of intravertebral and intergroup variation of controls was carried out using a Paired Two Sample T test. The 7.0 mm CCD screws restored pullout strength to 62% of control pullouts; 7.0 mm CD screws, to 85%; 7.0 mm VSP screws, to 99%; 8.0 mm CD screws, to 109%; and 8.0 mm VSP screws, to 148% of control pullouts. The 7.0 mm VSP salvage screws exceeded CD screws in ultimate load by 22.5% (p < .03) and CCD screws by 33.5% (p < .05). The 8.0 mm salvage screws significantly increased pullout relative to both controls and all 7.0 mm salvage screws, with 8.0 mm VSP exceeding 8.0 mm CD by 34% (p < .03). Initial stiffness was also significantly greater in VSP than CD screws. Although applied in a smaller number of vertebrae, 8.0 mm screws sufficiently outperformed smaller screws to provide statistically significant differences. The 7.0 mm VSP salvage screws restored pullout to control levels, roughly equivalent to outcomes previously obtained with unpressurized polymethylmethacrylate (PMMA).