Medical Student University of California, San Francisco San Francisco, California, United States
Introduction: Few studies have directly compared antepsoas versus transpsoas lateral lumbar interbody fusion (LLIF). We compared one-year radiographic and patient-reported outcomes after transpsoas versus antepsoas LLIF.
Methods: Patients undergoing LLIF without osteotomies were retrospectively reviewed at a single institution. Patient-reported outcomes – numeric rating scale (NRS) and EQ5D – were collected at one year postop. The average of the left and right foraminal height was calculated on preoperative and postoperative x-rays. The disc space was divided into thirds to characterize cage position. The antepsoas versus transpsoas approaches were compared utilizing univariate analysis. Because patients underwent multi-level fusions, a regression analysis that fits a between-effects model predicted the outcome (postoperative foraminal height).
Results: Sixty patients underwent 109 lateral interbody fusions. Mean EBL and operative time between the antepsoas (N=29 patients/ 50 levels) versus transpsoas approaches (N=31 patients/59 levels) did not differ. There was no difference in median number of levels fused (p=0.400). The antepsoas vs transpsoas approaches had similar rates of posterior decompressions (58.6% [N=17] vs 51.6% [N=16], p=0.586). The antepsoas versus transpsoas approaches did not differ with respect to postop NRS back pain (4.2 vs 4.7, p=0.516), NRS leg pain (3.5 vs 4.1, p=0.553), EQ5D (0.68 vs 0.64, p=0.564). Mean foraminal height improved in both the antepsoas (+2.2mm, p=0.002) and transpsoas (+3.6mm, p< 0.001) approaches; however, the postoperative foramen was higher in the transpsoas (18.0mm) versus antepsoas (16.1mm) approaches (p=0.014). The transpsoas approach had a significantly higher mean cage height (10.3mm vs 8.9mm, p< 0.001) and more anterior position of cage (p=0.001). When controlling for cage height and location of the cage, the lateral approach did not correlate with postoperative foraminal height restoration (p=0.177) following a regression analysis. Foraminal height correlated with NRS leg pain (p=0.019), but not NRS back pain or EQ5D.
Conclusion : Both the transpsoas and antepsoas approaches allow for a significant restoration in foraminal height. More importantly, attention should be paid to the position and size of the cage to improve foraminal height, which statistically significantly correlates with NRS leg pain.
