Pedicle screw augmentation in posterior constructs of the thoracolumbar spine: How many pedicle screws should be...

To evaluate the effects of different pedicle screw augmentation strategies on screw loosening and adjacent segment collapse at the proximal end of long-segment instrumentation.

Eighteen osteoporotic (9 male, 9 female donors; mean age: 74.7 ± 10.9 [SD] years) thoracolumbar multi-segmental motion segments (Th11 - L1) were assigned as follows: control, one-level augmented screws (marginally), and two-level augmented screws (fully augmented) groups (3 × 6). Pedicle screw placement was performed in Th12 and L1. Cyclic loading in flexion started with 100-500 N (4 Hz) and was increased by 5 N every 500 cycles. Standardized lateral fluoroscopy images with 7.5 Nm loading were obtained periodically during loading. The global alignment angle was measured to evaluate the overall alignment and proximal junctional kyphosis. The intra-instrumental angle was used to evaluate screw fixation.

Considering screw fixation as a failure criterion, the failure loads of the control (683 N), and marginally (858 N) and fully augmented (1050 N) constructs were significantly different (ANOVA p = 0.032).Taking the overall specimen alignment as failure criteria, failure loads of the three groups (control 933 ± 271.4 N, marginally 858 N ± 196 N, and full 933 ± 246.3 N were in the same range and did not show any significance (p = 0.825).

Global failure loads were comparable among the three groups and unchanged with augmentation because the adjacent segment and not the instrumentation failed first. Augmentation of all screws showed significant improved in screw anchorage.