Improved clinical and radiographic outcomes with expandable cages in transforaminal lumbar interbody fusion: a propensity-matched cohort analysis.

Abstract

Objective: The restoration of sufficient overall lumbar lordosis (LL) and segmental LL (SL) is associated with achieving optimal sagittal balance, decreasing back pain, and enhancing functional outcomes for patients. Expandable cages were developed in hopes of improving radiographic parameters and clinical outcomes, although current clinical evidence is inconclusive. Here, the authors aimed to evaluate the clinical and radiographic outcomes in patients undergoing one- or two-level open transforaminal lumbar interbody fusion (TLIF) with expandable versus static cage placement, using propensity-matched cohorts.

Methods: An institutional retrospective cohort of patients who underwent one- or two-level open TLIF with either expandable cage or static cage placement was identified. Using relevant preoperative covariates, including age, primary versus revision operation, number of cages implanted, and surgical level implanted, the authors built propensity-matched cohorts. They identified clinical outcomes in both cohorts, including operative characteristics and complication rates, along with pain, weakness, and sensory deficits over follow-up. Furthermore, they extracted and examined preoperative, postoperative, and last follow-up radiographic parameters.

Results: A total of 148 patients were included, and they were followed for a mean of 1.7 years (range 0.5-4.3 years). Propensity matching was used to create cohorts of patients who were similar with respect to age, surgical indication, revision status, number of cages implanted, surgical level implanted, and length of follow-up. Patients in both groups had similar preoperative radiographic parameters. Patients with expandable cages saw larger increases in SL, both postoperatively (5.3° ± 7.5° vs 1.6° ± 5.6°, p = 0.006) and at last follow-up (5.7° ± 7.4° vs 1.0° ± 6.1°, p = 0.003). They also saw significant improvements in pelvic incidence minus LL mismatch at last follow-up (-4.4° ± 13.2° vs 5.8° ± 13.8°, p = 0.009). No differences in intraoperative or perioperative complications were found, but patients with expandable cages were less likely to require readmission, develop adjacent-segment disease, or require revision surgery. They were also more likely to be symptom free at 1 month after surgery and at last follow-up.

Conclusions: Expandable cages lead to better restoration of radiographic features, including SL and improvements in clinical outcomes, compared with static cages in propensity-matched cohorts in patients undergoing one- or two-level open TLIFs.