ABSTRACT
BACKGROUND CONTEXT: The oblique prepsoas retroperitoneal approach to the lumbar spine for interbody fusion or oblique lumbar interbody fusion (OLIF) provides safe access to nearly all lumbar levels. A wide interval between the psoas and aorta allows for a safe and straightforward left-sided oblique approach to the discs above L5. Inclusion of L5-S1 in this approach, however, requires modifications in the technique to navigate the complex and variable vascular anatomy distal to the bifurcation of the great vessels. While different oblique approaches to L5-S1 have been described in the literature, to our knowledge, no previous study has provided guidance for the choice of technique. PURPOSE: Our objectives were to evaluate our early experience with the safety of including L5-S1 in OLIF using 3 different approach techniques, as well as to compare early complications between OLIF with and without L5-S1 inclusion. STUDY DESIGN: Retrospective cohort study. PATIENT SAMPLE: Of the 87 patients who underwent lumbar interbody fusion at 167 spinal levels via an OLIF approach, 19 included L5-S1 (group A) and 68 did not (group B). OUTCOME MEASURES: Demographics, levels fused, indications, operative time (ORT), estimated blood loss (EBL), vascular ligation, intraoperative blood transfusion, length of stay (LOS), discharge to rehabilitation facility, and complications (intraoperative, early ≤90 days, and delayed >90 days) were retrospectively assessed and compared between the groups. METHODS: A retrospective chart and imaging review of all consecutive patients who underwent OLIF at a single institution was performed. Indications for OLIF included symptomatic lumbar degenerative stenosis, deformity, and spondylolisthesis. The L5-S1 level, when included, was approached via one of the following 3 techniques: (1) a left-sided intrabifurcation approach; (2) left-sided prepsoas approach; and (3) right-sided prepsoas approach. Vascular anatomic variations at the lumbosacral junction were evaluated using the preoperative magnetic resonance imaging (MRI), and a "facet line" was proposed to assess this relationship. A minimum of 6 months of follow-up data were assessed for approach-related morbidities. RESULTS: Demographics and operative indications were similar between the groups. The mean follow-up was 10.8 (6-36) months. ORT was significantly longer in group A than in group B (322 vs. 256.3 min, respectively; p=.001); however, no difference in ORT between the two groups was found in the subanalyses for 2- and 3-level surgeries. Differences in EBL (260 vs. 207.91 cc, p=.251) and LOS (2.76 vs. 2.48 days, p=.491) did not reach statistical significance. Ligation of the iliolumbar vein, segmental veins, median sacral vessels, or any vascular structure, as needed for adequate exposure, was required in 13 (68.4%) patients from group A and 4 (5.9%) from group B (p<.00001). Two patients suffered minor vascular injuries (1 in each group); however, no major vascular injuries were seen. Complications were not significantly different between groups A and B, or between the three approaches to L5-S1, and trended lower in the latter part of the series as the learning curve progressed. CONCLUSIONS: Inclusion of L5-S1 in OLIF is safe and feasible through three different approaches but likely involves greater operative complexity. In our early experience, inclusion of L5-S1 showed no increase in early complications. This is the first series that reports the use of 3 different oblique approaches to L5-S1. The proposed "facet line" in the preoperative MRI may guide the choice of approach.
