Abstract
BACKGROUND: Deciding where to end a long-segment fusion for adult spinal deformity (ASD) may be a challenge, particularly in the absence of an abnormality at L5/S1. Some suggest prophylactic extension of the construct to the sacrum and/or ilium (S/I) to protect against distal junctional failure, while others support terminating in the lower lumbar spine to preserve motion.
OBJECTIVE: To compare the risk of re-operation after long-segment fusions for ASD that ends at L4 or L5 (L4/5) vs S/I.
METHODS: A multicenter database of patients treated for ASD by circumferential minimally invasive surgery or hybrid surgical technique was screened for individuals with long fusions (>= 4 vertebral levels) ending at L4 or below and with at least 2 yr of follow-up. Multivariate regression modeling was used to compare surgical morbidity between the L4/5 and S/I groups, and Cox proportional hazard modeling was used to compare risk of re-operation.
RESULTS: There were 45 subjects with fusion to L4/5 and 71 to S/I. Over a 32-mo median follow-up, 41 re-operations were performed; 6 were for distal junctional failure. In those with normal or mild degeneration at L5/S1, fusion to S/I afforded no significant change in re-operative risk (hazard ratio = 1.18 [ 95% confidence interval: 0.53-2.62], P = .682). In those undergoing circumferential minimally invasive surgery correction, fusion to S/I was associated with significantly greater blood loss (499.6 cc, P < .001) and surgical time (97.5 min, P = .04).
CONCLUSION: In the setting of a normal or mildly degenerated L5/S1 disc space, fusion to the sacrum/ilium did not significantly change the risk of requiring a re-operation after a long-segment fusion for ASD.
