Object: Of concern to spine surgeons are accelerated degenerative changes of motion segments located above and below where spinal fusion has been performed. Graf artificial ligament stabilization has been developed to avoid the adverse effect of spinal fusion. The object of this study was to assess the adjacent-segment morbidity of Graf ligamentoplasty compared with posterolateral fusion (PF) in which instrumentation was used.
Methods: Data obtained in 45 patients who underwent L4-5 Graf ligamentoplasty (18 patients) or PF with instrumentation (27 patients) were reviewed retrospectively. The minimum follow-up period was 5 years. In the PF group a solid fusion rate of 92.6% was achieved. Radiographic evaluation included assessment of lumbar sagittal alignment, range of motion (ROM), and adjacent-disc degeneration. Adjacent-segment morbidity was clinically assessed by determining the reoperation rate. Graf ligamentoplasty maintained regional lordosis and flexibility (13 degrees in L4-5 lordosis; 4.4 degrees in L4-5 ROM). Although there was no difference in preoperative adjacent-disc condition between the two groups, radiographic evidence of adjacent-disc deterioration was observed more frequently in patients in the PF group than the Graf group (25% and 6% at L1-2; 38% and 6% at L2-3; 38% and 18% at L3-4; and 43% and 18% at L5-sacrum, respectively). One case in the Graf group (5.6%) and five cases in the PF group (18.5%) required additional surgeries for adjacent-segment lesions.
Conclusions: Graf ligamentoplasty cannot completely replace spinal fusion. In a well-selected group of patients, however, it was shown to maintain lumbar mobility and sagittal alignment, and it decreased the risk of adjacent-segment deterioration compared with PF with instrumentation.