Anterior column realignment better restores sagittal alignment but carries higher risk of mechanical failures than lateral lumbar interbody fusion in patients with degenerative sagittal imbalance

BACKGROUND OF CONTEXT: Anterior column realignment (ACR), a modified lateral lumbar interbody fusion (LLIF), is an emerging, less invasive technique that allows greater lordosis correction by releasing anterior longitudinal ligament. However, long-term results have been poorly documented with regard to mechanical failure, such as proximal junctional kyphosis (PJK) and rod fracture (RF), and clinical outcomes. PURPOSE: To compare the outcomes, primarily mechanical failure, in patients with degenerative sagittal imbalance (DSI) treated with ACR versus LLIF alone. STUDY DESIGN/SETTING: Retrospective study. PATIENT SAMPLE: Patients ≥60 years of age; severe DSI defined by pelvic incidence (PI) - lumbar lordosis (LL) ≥20°; performance of ≥2-level LLIF; and ≥5 total fused levels including the sacrum. OUTCOME MEASURES: Mechanical failure such as PJK and RF; radiographic results; clinical outcomes METHODS: Enrolled patients were divided into two groups, based on whether their anterior reconstruction was accomplished with ACR or LLIF alone: ACR and LLIF groups. Mechanical failures were compared between the two groups as a composite outcome including PJK and /or RF. PJK was defined as proximal junctional angle (PJA) >28° and Δ PJA >22°. Only RFs developing at the level with corresponding procedures (ACR or LLIF) were included in the analysis. Logistic regression was performed to compare the relative risk of mechanical failure between the ACR and LLIF groups. The radiographic and clinical outcomes were also compared between the groups. RESULTS: The final study cohort consisted of 210 patients. The mean age was 69.6 years, and there were 190 females (90.5%). There were 124 patients in the ACR group and 86 patients in the LLIF group. Perioperative changes for all sagittal parameters were significantly greater in the ACR group than in the LLIF group. Overall mechanical failure rates were significantly higher in the ACR group than in the LLIF group (32.3% vs 14.0%, p=.003). Multivariate regression analysis with adjusting potential confounders revealed that ACR carried a significantly higher risk of mechanical failure than LLIF (Odds ratio=5.6, 95% confidence interval=2.0–15.6, p<.001). The final clinical outcomes were worse in the ACR group than in the LLIF group. CONCLUSION: ACR restored the sagittal malalignment more powerfully than did LLIF. However, compared to the LLIF, ACR was associated with a greater risk of mechanical failures and revision surgery. The final clinical outcomes in the ACR group were inferior to those in the LLIF group. Therefore, ACR should be left as a last resort for the cases where it is expected that an adequate correction cannot be achieved using LLIF alone. If ACR has to be performed, it is necessary to establish feasible surgical strategies to avoid mechanical failures.