Introduction. While pedicle subtraction osteotomies (PSO) in the lumbar spine are powerful techniques to correct sagittal and coronal spinal malalignment, they are fraught with high complication rates, including nonunions and rod fractures due to motion at the PSO site. Multi-rod constructs are used commonly to decrease rates of pseudarthrosis and can be achieved with "satellite" rods (not connected to primary rod) placed in-line or laterally to the primary rod. This study aimed to assess the biomechanical properties of laterally-based satellite rods as well as evaluate effects of screw-type (polyaxial v. monoaxial) spanning the PSO.

Materials and Methods. A validated 3D spinopelvic finite element model with a 30° PSO at L3 was used. To evaluate 5 models: (1) Control (T10-pelvis+2 rods); (2) lateral satellite rods connected via offsets to monoaxial screws (LatSat-Mono) or (3) polyaxial screws (LatSat-Poly); (4) in-line satellite rods connected to monoaxial screws (InSat-Mono) or (4) polyaxial screws (InSat-Poly) (Fig. 1). Global and PSO range of motion (ROM) were recorded. Rods’ von Mises stresses and PSO forces were recorded and the percent differences from Control were calculated.

Results. Multi-rods constructs with satellite rods in any configuration (in-line and laterally) decreased global ROM in flexion, extension and lateral bending. Compared to the control, PSO flexion and extension ROM increased for all satellite rods, except for InSat-Mono that decreased flexion ROM at the PSO.  Laterally based satellite rods increased PSO forces (347.1N in LatSat-Mono, 348.6N in LatSat-Poly) compared to control (336N). Conversely, in-line satellite rods decreased PSO forces (280.1N in InSat-Mono, and 330.7N in InSat-Poly).  Locations of maximal von-Mises stress on the primary rods were adjacent to the PSO site and L5-S1 (2).  Both satellite rod configurations (in-line and lateral) decreased von-Mises stresses on the primary rods at the PSO site. Von-Mises stress on satellite rods were lower on laterally-based satellite rods than in-line satellite rods. There were relatively high stresses on “satellite rods in the InSat-Mono model.

Discussion. Multi-rod constructs using in-line and lateral satellite rods across a lumbar PSO reduced stresses on primary rods via different mechanisms. Lateral satellite rods, irrespective of screw type, provided more favorable biomechanics, as they increased PSO forces and had lower von Mises stresses (lower posterior instrumentation stress shielding) compared to in-line satellite rods that had a high degree of posterior instrumentation stress shielding and lower PSO forces. Additional clinical studies are warranted to confirm these biomechanical findings.

Figure 1