Lateral lumbar interbody fusion at L4-5: a morphometric analysis of psoas anatomy and cage placement

Background: The ventral elevation (VE) of the psoas from the vertebral column as seen on axial magnetic resonance imaging has been suggested as a preclusion of a safe working zone for lateral lumbar interbody fusion (LLIF) at L4-5. However, no quantitative study has been conducted to verify this. Methods: L4-5 LLIF was attempted using 22-mm wider cages whenever feasible over 18-mm standard cages in a consecutive series of 62 patients. The degree of VE of the psoas measured from the anterior border of the L5 endplate and the anteroposterior cage position measured from the posterior vertebral border to the center of the cage normalized to the anteroposterior width of the L5 endplate were recorded on axial images. Results: LLIF was successfully performed in all cases despite 18 patients (29%) harboring elevated psoas. Standard cage was applied in 25 (40%) patients due to nerve proximity. Wider cage was applied equally frequent in both elevated (56%) and nonelevated groups (60%, P = 0.78). The mean cage position was found to be 0.54 ± 0.10 (standard deviation) and no difference was detected between the elevated (0.54 ± 0.11) and nonelevated (0.53 ± 0.10, P = 0.78) groups. Multivariate analyses and receiver operating characteristic analysis demonstrated that VE displayed low performance in predicting the choice of cages. Conclusions: LLIF is feasible in patients with ventrally elevated psoas, and the choice of cages does not appear to be influenced by the location of the psoas. Relying on the rising psoas sign on magnetic resonance imaging as a case selection criterion may unjustly exclude patients from LLIF.