Predictors of indirect neural decompression in minimally invasive transpsoas lateral lumbar interbody fusion

Objective: An advantage of lateral lumbar interbody fusion (LLIF) surgery is the indirect decompression of the neural elements that occurs because of the resulting disc height restoration, spinal realignment, and ligamentotaxis. The degree to which indirect decompression occurs varies; no method exists for effectively predicting which patients will respond. In this study, the authors identify preoperative predictive factors of indirect decompression of the central canal.

Methods: The authors performed a retrospective evaluation of prospectively collected consecutive patients at a single institution who were treated with LLIF without direct decompression. Preoperative and postoperative MRI was used to grade central canal stenosis, and 3D volumetric reconstructions were used to measure changes in the central canal area (CCA). Multivariate regression was used to identify predictive variables correlated with radiographic increases in the CCA and clinically successful improvement in visual analog scale (VAS) leg pain scores.

Results: One hundred seven levels were treated in 73 patients (mean age 68 years). The CCA increased 54% from a mean of 0.96 cm2 to a mean of 1.49 cm2 (p < 0.001). Increases in anterior disc height (74%), posterior disc height (81%), right (25%) and left (22%) foraminal heights, and right (12%) and left (15%) foraminal widths, and reduction of spondylolisthesis (67%) (all p < 0.001) were noted. Multivariate evaluation of predictive variables identified that preoperative spondylolisthesis (p < 0.001), reduced posterior disc height (p = 0.004), and lower body mass index (p = 0.042) were independently associated with radiographic increase in the CCA. Thirty-two patients were treated at a single level and had moderate or severe central stenosis preoperatively. Significant improvements in Oswestry Disability Index and VAS back and leg pain scores were seen in these patients (all p < 0.05). Twenty-five (78%) patients achieved the minimum clinically important difference in VAS leg pain scores, with only 2 (6%) patients requiring direct decompression postoperatively due to persistent symptoms and stenosis. Only increased anterior disc height was predictive of clinical failure to achieve the minimum clinically important difference.

Conclusions: LLIF successfully achieves indirect decompression of the CCA, even in patients with substantial central stenosis. Low body mass index, preoperative spondylolisthesis, and disc height collapse appear to be most predictive of successful indirect decompression. Patients with preserved disc height but severe preoperative stenosis are at higher risk of failure to improve clinically.