Lumbosacral Interbody Support is Not Necessary In Long Adult Deformity Constructs with Iliac Screw Fixation
Interbody support has been advocated for long deformity constructs in order to combat high non-union rates at the lumbosacral junction. A limited number of recent studies have suggested that large amounts of rhBMP-2 can be used in lieu of interbody fusion to obtain similar outcomes. No study to date has examined whether strictly refraining from anterior column fusion at L5-S1 alone without increasing rhBMP-2 use results in any functional or radiographic outcome changes.
To determine whether additional interbody fusion at L5-S1 leads to lower non-union and revision rates compared with posterior fusion and instrumentation alone in long deformity constructs to the pelvis.
Retrospective cohort analysis of prospectively collected data from 2 surgeons at a single institution.
95 consecutive patients with >2 year follow-up who underwent spinal fusion from T10 or higher down to the ilium for adult deformity (n=49 degenerative scoliosis, n=45 idiopathic, n=1 neuromuscular).
Radiographic measurements, Visual-analog pain scale (VAS), Oswestry Disability Index (ODI), and complications
68 patients who received a L5-S1 interbody cage with 4mg of bmp (group IB) were compared with 27 patients who did not receive interbody fusion or anterior column BMP at L5-S1 (group NI). All patients had fusion from T10 or higher down to the pelvis, which was secured with unilateral or bilateral iliac screw fixation. The 2 groups were similar in terms of number of levels fused, amount of BMP used posteriorly, and pre-op VAS scores. The NI group had significantly more previous surgeries (55.6% vs. 22.1%, p=0.014) and inferior pre-op ODI scores (50 vs 44, p=0.0045).
At 2 years post-op, both groups had similar improvement in coronal and sagittal alignment. Both groups also had significant but comparable improvement in clinical function at 2 years post-op (?VAS NI: -3.4, IB:-3.0 and ? ODI NI: -27, IB: -22). There were no substantial differences in complications overall, including non-union (NI: 1[3.7%] vs IB: 4[5.9%]), infection (NI: 0[0%] vs IB: 2[2.9%]), spinal imbalance (NI: 2[7.4%] vs IB: 2[2.9%]), and revision surgery (NI: 5[18.5%] vs IB: 11[16.2%]). Medical complication rates were also similar.
Additional L5-S1 interbody fusion does not appear to yield significant clinical or radiographic outcome improvement in adult deformity patients undergoing long fusion to the ilium using modern instrumentation techniques. Complication rates are also not significantly different with the addition of a lumbosacral cage, although costs are higher.