Bone graft volumetric changes and clinical outcomes after instrumented lumbar or lumbosacral fusion: a prospective cohort study with a five-year follow-up.

STUDY
DESIGN: Prospective cohort study.
OBJECTIVE: To determine: (1) patterns of bone graft volumetric changes, (2) whether different fusion methods, numbers of fusion segments, and range of fusion affect bone graft volumetric changes, and (3) whether these changes affect clinical and functional outcomes after surgery. SUMMARY OF BACKGROUND DATA: Bone graft volumetric change is likely to importantly determine successful fusion. However, despite the importance of graft volumetric changes, little information is available on bone graft volumetric changes after instrumented fusion.
METHODS: Thirty-one patients with lumbosacral disease who underwent decompression and instrumented spinal fusion using autologous iliac bone graft were included in this study. Patients were classified into 3 major categories, based on fusion methods, the number of fusion segments, and range of fusion, namely, (1) the posterolateral fusion and the posterior lumbar interbody fusion groups, (2) the one-segment fusion and 2-segment fusion groups, and (3) the floating fusion (fusion to L5) and fixed fusion (fusion to S1) groups. To evaluate bone graft volumetric changes, computed tomography scans were performed. Group graft volumes were compared, and Oswestry Disability Index and visual analogue scales scores were obtained to evaluate clinical and functional outcomes.
RESULTS: Initial graft volumes had decreased by 8% at 6 months and by 26.3% between 6 and 12 months after surgery (P = 0.0025, P = 0.0001, respectively). Mean group bone graft volume losses at 5 years after surgery were not significantly different between the posterolateral fusion and posterior lumbar interbody fusion groups or the floating fusion and fixed fusion groups (P = 0.722, P = 0.076, respectively). However, mean bone graft volume loss was significant greater in the 2-segment fusion group than in the one-segment fusion group (P = 0.007). No statistical difference in visual analogue scales and Oswestry Disability Index score decreases were observed between the groups. Mean graft bone volume loss during 5 years increased with initial graft volume (r = 0.525, P = 0.003). However, ratios of volume loss over 5 years versus initial graft volume decreased with initial graft volume (r = -0.459, P = 0.011).
CONCLUSION: One-third of initial bone grafts were found to be resorbed over 5 years and the majority of this resorption was found to occur between 6 months and 1 year after instrumented lumbar/lumbosacral fusion. Furthermore, bone graft volumetric changes were not affected by fusion methods or ranges of fusion, and did not affect clinical or functional outcomes. Although bone graft volume loss was found to increase with initial graft volume, larger bone grafts were found to result in a larger fusion mass.