Implant Density at the Apex Is More Important Than Overall Implant Density for 3D Correction in Thoracic Adolescent Idiopathic Scoliosis Using Rod Derotation and En Bloc Vertebral Derotation Technique

Study Design. Biomechanical analysis of 3D correction and bone-screw forces through numerical simulations of scoliosis instrumentation with different pedicle screw patterns. Objective. To analyze the effect of different screw densities and distributions on 3D correction and bone-screw forces in adolescent idiopathic scoliosis (AIS) instrumentation. Summary of Background Data. Instrumentation constructs with various numbers of pedicle screws and patterns have been proposed for thoracic AIS instrumentation. However, systematic biomechanical studies have not yet been completed on the appropriate screw patterns for optimal 3D correction. Methods. Patient-specific biomechanical models of the spine were created for 10 AIS cases (Lenke 1). For each case, surgical instrumentation patterns were computationally simulated using respectively a reference screw pattern (two screws per level fused) and six alternative screw patterns with fewer screws. Simulated surgical maneuvers and model definition were unchanged between simulations except the number and distribution of screws. 3D correction and bone-screw forces were compared. Results. A total of 140 posterior instrumentations were computationally simulated. Mean corrections in the coronal and sagittal planes with alternative screw patterns were within 48 to the reference pattern. Increasing screw density in the apical region from one to two screws per level improved percent apical vertebral rotation (AVR) correction (r = 0.887, P < 0.05). Average bone-screw force associated with the reference screw pattern was 243N +/- 54N and those with the alternative screw patterns were 11% to 48% lower. Conclusion. Compared with the reference maximal screw density pattern, alternative screw patterns allowed similar corrections in the coronal and sagittal planes. AVR correction was strongly correlated with screw density in the apical region; AVR correction varied significantly with screw patterns of the same overall screw density when an en bloc vertebral derotation technique was simulated. High screw density tended to over-constrain the instrumented spine and resulted in higher forces at the bone-screw interface.