Comparison of local and regional radiographic outcomes in minimally invasive and open TLIF: a propensity score–matched cohort

Abstract

OBJECTIVE Local and regional radiographic outcomes following minimally invasive (MI) transforaminal lumbar interbody fusion (TLIF) versus open TLIF remain unclear. The purpose of this study was to provide a comprehensive assessment of local and regional radiographic parameters following MI-TLIF and open TLIF. The authors hypothesized that open TLIF provides greater segmental and global lordosis correction than MI-TLIF. METHODS A single-center retrospective cohort study of consecutive patients undergoing MI- or open TLIF for grade I degenerative spondylolisthesis was performed. One-to-one nearest-neighbor propensity score matching (PSM) was used to match patients who underwent open TLIF to those who underwent MI-TLIF. Sagittal segmental radiographic measures included segmental lordosis (SL), anterior disc height (ADH), posterior disc height (PDH), foraminal height (FH), percent spondylolisthesis, and cage position. Lumbopelvic radiographic parameters included overall lumbar lordosis (LL), pelvic incidence (PI)–lumbar lordosis (PI-LL) mismatch, sacral slope (SS), and pelvic tilt (PT). Change in segmental or overall lordosis after surgery was considered "lordosing" if the change was > 0° and "kyphosing" if it was ≤ 0°. Student t-tests or Wilcoxon rank-sum tests were used to compare outcomes between MI-TLIF and open-TLIF groups. RESULTS A total of 267 patients were included in the study, 114 (43%) who underwent MI-TLIF and 153 (57%) who underwent open TLIF, with an average follow-up of 56.6 weeks (SD 23.5 weeks). After PSM, there were 75 patients in each group. At the latest follow-up both MI- and open-TLIF patients experienced significant improvements in assessment scores obtained with the Oswestry Disability Index (ODI) and the numeric rating scale for low-back pain (NRS-BP), without significant differences between groups (p > 0.05). Both MI- and open-TLIF patients experienced significant improvements in SL, ADH, and percent corrected spondylolisthesis compared to baseline (p < 0.001). However, the MI-TLIF group experienced significantly larger magnitudes of correction with respect to these metrics (ΔSL 4.14° ± 4.35° vs 1.15° ± 3.88°, p < 0.001; ΔADH 4.25 ± 3.68 vs 1.41 ± 3.77 mm, p < 0.001; percent corrected spondylolisthesis: −10.82% ± 6.47% vs −5.87% ± 8.32%, p < 0.001). In the MI-TLIF group, LL improved in 44% (0.3° ± 8.5°) of the cases, compared to 48% (0.9° ± 6.4°) of the cases in the open-TLIF group (p > 0.05). Stratification by operative technique (unilateral vs bilateral facetectomy) and by interbody device (static vs expandable) did not yield statistically significant differences (p > 0.05). CONCLUSIONS Both MI- and open-TLIF patients experienced significant improvements in patient-reported outcome (PRO) measures and local radiographic parameters, with neutral effects on regional alignment. Surprisingly, in our cohort, change in SL was significantly greater in MI-TLIF patients, perhaps reflecting the effect of operative techniques, technological innovations, and the preservation of the posterior tension band. Taking these results together, no significant overall differences in LL between groups were demonstrated, which suggests that MI-TLIF is comparable to open approaches in providing radiographic correction after surgery. These findings suggest that alignment targets can be achieved by either MI- or open-TLIF approaches, highlighting the importance of surgeon attention to these variables.