Introduction: Expandable interbody cages have evolved since their introduction to the market in the early 2000s, with the goal of reducing rates of cage migration, expulsion, nonunion, and subsidence¹. Notably, larger contact area has been shown to require significantly greater force increments to induce clinically significant subsidence². However, data describing contact area of modern expandable TLIF cages is limited. Interest in this data has grown as surgeons and device manufacturers increasingly seek to incorporate expandable cage technology with minimally invasive techniques. The objective of the present study was to evaluate cross-sectional surface area (CSA) of modern expandable TLIF cages and explore relationships between cage footprints and their respective torque-out value, maximum lordosis, and expansion heights.

Methods: Publicly available FDA, patent information, and product technique guides were evaluated for pre- and post-expansion cage length, width, height, lordosis, and torque-out value for the smallest and largest cages manufactured. Torque-out value was treated as a surrogate for force generated at maximum expansion. CSA of rectangular and banana-shaped cages were calculated as the CSA of a rectangle or an ellipse, respectively. Pearson’s Correlation and regression were used to evaluate for relationships between CSA and the above parameters.

Results: Twenty-one expandable TLIF cages were identified (two banana), with CSA ranging from 176-630mm². Lordosis ranged from 0-23°. Torque-out values ranged from 10-40 in-lbs. Maximum cage expansion ranged from 3-10mm of total excursion. Of the smallest and largest cages offered, collapsed to expanded cage heights ranged from 6-17mm and 7-17mm, respectively.

Of the largest cages offered for a specific device, CSA positively correlated with maximally expanded height (r=0.497, p=0.042) and maximum range-of-expansion (r=0.617, p=0.008). There were no significant relationships between CSA and lordosis, torque-out value, or collapsed height. Cage-shape did not predict torque-out value (p=0.366, OR=1.087, CI⁹⁵=0.907-1.300).

The mean ratios of CSA-to-collapsed-height and of CSA-to-expanded-height were 31mm and 19mm for the smallest cages offered, respectively, and significantly differed (p<0.001), and were 33mm and 21mm for the largest cages offered, respectively, and significantly differed (p<0.001).

Conclusions: The smallest and largest cages had similar collapsed to maximally-expanded heights. Amongst the largest sized cages offered, cages with larger CSA had greater maximally expanded cage height and total expansion range, but the ratio of proportional CSA per expansion was significantly lower at maximally expanded height compared to fully collapsed height. Thus, expandable cages had less proportional CSA at maximum expansion. Further research evaluating CSA and endplate-contact-force would further highlight the relationship between parameters associated with subsidence within expandable cages.