Rod fractures and nonunions after long fusion to the sacrum for primary presentation adult spinal deformity: a comparison with and without interbody fusion in the distal lumbar spine

Study design: Retrospective cohort study.

Objectives: To investigate the prevalence and incidence rate of rod fractures (RF) in patients undergoing surgery for correction of adult spinal deformity (ASD) with or without the use of interbody fusions in the caudal levels of the fusion construct.

Background: Data: Pseudarthrosis and rod fracture after long spinal fusion to the sacrum for correction of ASD remain a concern.

Methods: We reviewed clinical records of patients who underwent surgery for correction of ASD between 2004 and 2014. All cases were primary (no prior spine fusion) surgeries with long fusion to the sacrum and bilateral spinopelvic fixation. Patients were dichotomized into one of two groups based on whether an interbody fusion was performed at the caudal levels of the fusion construct. The primary outcome of interest was the prevalence and incidence rate of RFs.

Results: A total of 230 patients underwent a long segment fusion for correction of ASD with mean follow-up of 55 months. 117 patients had an interbody fusion (IF) while 113 patients did not (NIF). At last follow-up, there was no significant difference in the prevalence of RFs between the cohort of patients IF vs NIF (IF cohort: n = 20, 17.9% vs NIF cohort: n = 15, 14.2%, p = 0.49). However, the incidence rate for bilateral rod fractures was 1.6%/year for IF group vs 1.0%/year for NIF group (p = 0.02). Location of RF was different between the two groups; RF (unilateral and bilateral) above L4 was the most common location in the IF group (n = 17/20; 85%) compared to L4-S1 in the NIF group (n = 11/15; 73%) (p = 0.02).

Conclusion: Interbody fusion does not fully protect against rod failure in the lumbar spine in ASD patients with long posterior spinal fusion and may encourage failure at L2-L4, the levels above the interbody fusion.

Level of evidence: III.