Biomechanical comparison of single-level posterior versus transforaminal lumbar interbody fusions with bilateral pedicle screw fixation: segmental stability and the effects on adjacent motion segments

Object Both posterior lumbar interbody fusion (PLIF) and transforaminal lumbar interbody fusion (TLIF) have been frequently undertaken for lumbar arthrodesis. These procedures use different approaches and cage designs, each of which could affect spine stability, even after the addition of posterior pedicle screw fixation. The objectives of this biomechanical study were to compare PLIF and TLIF, each accompanied by bilateral pedicle screw fixation, with regard to the stability of the fused and adjacent segments. Methods Fourteen human L2–S2 cadaveric spine specimens were tested for 6 different modes of motion: flexion, extension, right and left lateral bending, and right and left axial rotation using a load control protocol (LCP). The LCP for each mode of motion utilized moments up to 8.0 Nm at a rate of 0.5 Nm/second with the application of a constant compression follower preload of 400 N. All 14 specimens were tested in the intact state. The specimens were then divided equally into PLIF and TLIF conditions. In the PLIF Group, a bilateral L4–5 partial facetectomy was followed by discectomy and a single-level fusion procedure. In the TLIF Group, a unilateral L4–5 complete facetectomy was performed (and followed by the discectomy and single-level fusion procedure). In the TLIF Group, the implants were initially positioned inside the disc space posteriorly (TLIF-P) and the specimens were tested; the implants were then positioned anteriorly (TLIF-A) and the specimens were retested. All specimens were evaluated at the reconstructed and adjacent segments for range of motion (ROM) and at the adjacent segments for intradiscal pressure (IDP), and laminar strain. Results At the reconstructed segment, both the PLIF and the TLIF specimens had significantly lower ROMs compared with those for the intact state (p < 0.05). For lateral bending, the PLIF resulted in a marked decrease in ROM that was statistically significantly greater than that found after TLIF (p < 0.05). In flexion-extension and rotation, the PLIF Group also had less ROM, however, unlike the difference in lateral bending ROM, these differences in ROM values were not statistically significant. Variations in the position of the implants within the disc space were not associated with any significant differences in ROM values (p = 0.43). Analyses of ROM at the adjacent levels L2–3, L3–4, and L5–S1 showed that ROM was increased to some degree in all directions. When compared with that of intact specimens, the ROMs were increased to a statistically significant degree at all adjacent segments in flexion-extension loads (p < 0.05); however, the differences in values among the various operative procedures were not statistically significant. The IDP and facet contact force for the adjacent L3–4 and L5–S1 levels were also increased, but these values were not statistically significantly increased from those for the intact spine (p > 0.05). Conclusions Regarding stability, PLIF provides a higher immediate stability compared with that of TLIF, especially in lateral bending. Based on our findings, however, PLIF and TLIF, each with posterolateral fusions, have similar biomechanical properties regarding ROM, IDP, and laminar strain at the adjacent segments.