Lumbar spondylolisthesis is a degenerative change of the lumbar spine with typical clinical manifestations. For patients who have not succeeded in conservative treatment, lumbar fusion surgery is one of the treatment options available. Lumbar fusion procedures, including anterior lumbar interbody fusion (ALIF), posterior lumbar interbody fusion (PLIF) and direct lateral interbody fusion (DLIF), all have their own unique advantages and inevitable risks. It is important that each patient's anatomy, age, expectations, concerns and the experience of the surgeon are factored into the decision-making process to determine the best strategy for the individual patient to maximize recovery while minimizing the risk of complications. In the OLIF procedure, the posterior column structure is preserved, the imbalance between the coronal and sagittal planes is corrected and excellent clinical results have also been demonstrated.[21, 22] Furthermore, the efficacy of MIS-TLIF has been proven and is a better choice for patients.[23–26]
In our present retrospective study, we first compared some intraoperative and postoperative data between the two groups. The OLIF group outperformed the MIS-TLIF group in terms of operative time, intraoperative bleeding and postoperative drainage(Table 1). The OLIF procedure reduces operative time and intraoperative injury by bluntly separating the natural corridor from the artery to the psoas major muscle to access the intervertebral space, rather than successive dilatation of the paravertebral muscle and subtotal joint resection as in MIS-TLIF. Furthermore, the indirect decompression of OLIF reduces the time spent using intraoperative neuromonitoring. Although these indicators may related to operator technique and individual anatomical differences, we prefer to believe that OLIF has better perioperative results. Surely, the steep learning curve of MIS-TLIF is also a relevant factor. Only single-segment fusion was studied in the current study, we theoretically believe that OLIF has an advantage when performing multi-stage fusion, which can be done via a single channel. Additionally, there was a significant difference in VAS between the two groups at three days postoperatively, which may be due to the fact that OLIF has no effect on the paravertebral muscles. Neither ODI nor VAS were significantly different at subsequent follow-up. The availability of MED for nerve root decompression under direct vision and the higher intervertebral space recovery in the OLIF group both contributed to the improvement of the patient's symptoms. The MED-assisted MIS-TLIF maximizes anatomical preservation, allows for the re-treatment of the nucleus pulposus under direct vision, preserves the ligamentum flavum, and allows for the placement of screws under MED access with minimal trauma and rapid recovery and guaranteed clinical efficacy.
Changes in the physiological curvature of the spine and loss of DH have both been found to contribute to lower back pain in patients and it is clinically important to study their changes. In our current retrospective study, there was a statistically significant increase in imaging parameters in both groups compared to the preoperative period and the degree of improvement was statistically significant at each subsequent follow-up. (Table 4,6). The OLIF procedure improves the patient's symptoms by increasing the disc height through the placement of a larger cage, lengthening the hypertrophic ligamentum flavum and achieving indirect nerve decompression. The larger DH improvement in the OLIF group was attributed to the lateral surgical access allowing for the placement of a larger cage, which was not possible in the MIS-TLIF group due to tissue obstruction and channel limitations. This is consistent with previous reports. The Ko, M. J.  study showed that placement in the strongest position at the front of the endplate helped restore lumbar lordosis. The MIS-TLIF group in the study did not have a significant effect on the recovery of sagittal angulation and the relevant factors, in addition to the size and position of the cage, the fusion of a single segment may have prevented it from performing excellently.
The incidence of CS and fusion are related to factors such as damage to the endplate, material of the cage, and bone quality. Higher magnitudes of CS were associated with worse surgical improvements.  Hence, CS should be considered an early postoperative complication, so delicate and gentle intraoperative manipulation, long-term postoperative intervention is necessary. In our study, neither postoperative CS nor fusion rates were statistically significant at the final follow-up between the two groups(P = 0.775, 0.206), but were both better than in previous studies. Such a result may be associated with our better intraoperative protection of the bony endplate, the addition of additional screw fixation and shorter duration of follow-up. Biomechanics shows that the stress load on the fixator in OLIF is less than that of the MIS-TLIF group, which is more advantageous in terms of postoperative CS and fusion rates.[36, 37] One study divided the vertebral body into five zones from anterior to posterior. When the larger Cage is placed in the more rigid Zone II, the cage spanning the entire width of the vertebral body reduces the stress distribution in the vertebral endplate and cancellous bone, and also increases the maximum stress load. This facilitates improved fusion rates and resistance to subsidence. This has contributed to the better performance of OLIF, whereas MIS-TLIF cage are mostly placed near the center of the vertebral endplate.[39, 40] Therefore, we have reason to believe that the advantages of OLIF will become increasingly apparent after a long follow-up period.
Different surgical approaches can lead to different complications. The OLIF creating access through the natural gap between the psoas major muscle and the abdominal vessels, avoiding the need to expose dura and thereby reduce the chances of dural injury. The MIS-TLIF in our study is based on the MED by creating a working channel through the Wiltse approach to reach the surgical area, requiring only partial removal of the articular processes and the vertebral plates, avoiding unnecessary damage to the multifidus muscle, major vessels like aorta and better maintaining spinal stability.
In our present retrospective study, three patients in the OLIF group suffered from numbness of the thighs and weakness of the psoas major muscle after surgery all of which resolved at follow-up, another patient presented with ureteral injury. These are related to the anatomy of the body. The psoas major muscle mainly originates from the diaphragmatic fascia and the front of the transverse processes, vertebrae and discs of the L1-L5 lumbar vertebrae, ends at the lesser trochanter of the femur and the fascia of the pelvic floor. We consider that excessive contraction of the psoas major may lead to postoperative numbness in the groin, anterior thighs or weakness of the psoas major. When we choose an inappropriate surgical incision, the surgical access is too vertical and the duration of the operation is too long it may cause symptoms of weakness of the psoas major. The ureter is located retroperitoneally and passes anteriorly down into the pelvis through the medial aspect of the psoas major muscle. Complete retrieval of the retroperitoneal adipose tissue before initiating disc removal can avoid urinary tract injury. The ureter should be repaired immediately after the diagnosis is clear. A life-threatening complication of OLIF is the intraoperative rupture of a large vessel, the incidence of which has been reported to be 0.3%-2.4%. However, this did not occur intraoperatively, which was related to precise preoperative planning, and we performed a preoperative CTA of the abdomen in each patient to exclude individual anatomical variation. In the MIS-TLIF group, A patient with a postoperative cerebrospinal fluid leak due to a ruptured dural sac was lying flat and resting for three days before no more cerebrospinal fluid was seen in the drainage bag and the dizziness subsided. Another patient presented with a hematoma. Postoperative nerve root symptoms and rupture of the dural sac are difficult to avoid. MIS-TLIF is a direct decompression of the nerve roots, and our posterior access to the disc necessarily affects the paravertebral tissues and requires partial resection of the intervertebral joints.
This study is also limited in that it is a non-randomized, retrospective single-center study. This study only included single-level lumbar fusion. Clinical outcomes, radiological results and complications may differ when multi-segment procedures are performed. In addition, this study also had limitations, with only a short follow-up period. Longer-term follow-up and more high-quality randomized controlled trials are needed to confirm the results of the current study.