Background
Lumbar interbody fusion is effective in enhancing arthrodesis, inducing lumbar lordosis and decompressing neural elements [
1‐
3]; therefore, it has been widely used as the gold-standard treatment for a variety of end-stage lumbar degenerative disorders, including stenosis, disc disease, spondylolisthesis and deformity [
4]. Studies have validated the therapeutic effect of conventional open posterior or anterolateral surgeries [
5,
6]. However, due to significant iatrogenic injury and approach-related morbidity, these surgeries have been gradually replaced by newly developed minimally invasive procedures [
7,
8].
As an alternative to conventional posterior-approach surgery [
9], minimally invasive transforaminal lumbar interbody fusion (MIS-TLIF) not only reduces iatrogenic soft tissue injury during spinal exposure [
10,
11] but also minimizes the retraction of the dural sac and nerve root through its transforaminal corridor [
12,
13]. In contrast, oblique lateral interbody fusion (OLIF) is a revised type of anterolateral-approach technique that uses the retroperitoneal corridor to access the intervertebral space and decompresses the neural structure in an indirect manner [
14,
15]. OLIF effectively avoids hypogastric or lumbar plexus injuries, and it keeps the psoas intact. In addition, OLIF preserves the posterior structures [
16,
17]. To date, both MIS-TLIF and OLIF are steadily gaining acceptance as the choice of fusion methods as their clinical and radiographic efficacy has been demonstrated by a growing body of evidence [
9,
18‐
23].
Nevertheless, due to the distinct surgical approach and fusion strategy of MIS-TLIF and OLIF, it remains to be determined which procedure leads to better outcomes. To our knowledge, only a few studies have compared them directly, and convincing evidence is still lacking. In this context, a meta-analysis was performed to compare the clinical and radiographic outcomes of MIS-TLIF and OLIF to provide an evidence-based reference for clinicians.
Discussion
In recent decades, both MIS-TLIF and OLIF have become the most widely used minimally invasive techniques in the treatment of lumbar degenerative diseases, but their relative efficacy has yet to be determined. By systematically comparing their outcomes in single-level interbody fusion, our analysis suggested that OLIF remarkably reduces the operation time (with supplementary posterior fixation in the prone position) and intraoperative blood loss compared with MIS-TLIF. This improvement may be attributed to the delicate retroperitoneal approach in OLIF, where the surgeons access the intervertebral spaces through the natural corridor between arteries and psoas muscles by simple blunt dissection instead of successive paraspinal muscle dilation and facetectomy as in MIS-TLIF [
23,
33]. Moreover, direct decompression is avoided in OLIF, which largely eliminates the additional time occupied by microscopy and other specific complex ancillaries for neuroprotection or monitoring [
14,
34]. In three Japanese studies, surgeons attempted to accomplish posterior fixation in the same lateral position to reduce the additional operation time caused by position changes in OLIF [
30‐
32]. Unexpectedly, the operation time did not decrease and was similar to that of MIS-TLIF, which might be due to the limited working space and difficulty in fluoroscopy at this position. In the present study, we only analysed single-level interbody fusion. In fact, OLIF achieves multilevel fusion through the same corridor, while separate channels are required in MIS-TLIF. Theoretically, we assume that the superiority of OLIF in operation time and intraoperative blood loss may further expand with the increase in the number of fusion segments [
35,
36].
Despite being less invasive, it remains unclear whether OLIF attains similar clinical effects to MIS-TLIF. Unlike MIS-TLIF, OLIF only achieves neural decompression indirectly through obliteration of the bulging disc and ligamentotaxis [
37]. In fact, our pooled analysis indicated that OLIF may be even more effective in alleviating leg pain, which agreed with the imaging finding that a significantly increased cross-sectional area of the foramen is observed in patients undergoing OLIF rather than MIS-TLIF, suggesting a better performance of OLIF in nerve decompression [
26]. Additionally, as OLIF does not open the spinal canal, stimulation to nerve roots is avoided, which may further alleviate leg symptoms after surgery [
38]. Although without statistical significance, OLIF still showed an advantage in reducing pain in the back. Some surgeons have proposed that this advantage may be attributed to the better protection of lumbar fascia and paravertebral muscles in OLIF [
31]. In terms of functional recovery, improvement of the ODI scores was similar between MIS-TLIF and OLIF. A similar efficacy was also confirmed by the JOABPEQ effectiveness rate regarding lumbar function, walking ability, social life and mental health. Only term associated with pain was in line with the VAS score, which favoured OLIF.
With respect to complications, the overall complication incidence of MIS-TLIF (13.8 %) was slightly less than that of OLIF (16.3 %); however, the difference was statistically nonsignificant. Attributed to the narrow transforaminal corridor, dural tears and nerve root injuries are the most common approach-related complications in MIS-TLIF [
11,
39,
40]. In contrast, thigh numbness along with hip flexor weakness are frequent after OLIF, which might be due to damage to psoas nerve branches and continuous psoas traction [
41‐
43]. Usually, most symptoms disappear in three months [
44]. Approach-unrelated complications, including deep venous thrombosis, infection and pulmonary thromboembolism, were comparable between MIS-TLIF and OLIF. However, at the same time, some studies have also revealed a shorter bedridden time after OLIF, suggesting that it may be beneficial for the rehabilitation and reduction of bedridden complications [
45].
Radiographically, pooled analysis demonstrated that OLIF led to better disc height improvement than MIS-TLIF [
26,
42], which may be due to the generation of a significantly large annular window in OLIF for sufficient anterior release and high-profile cage implantation [
46,
47]. Due to the obstacles of the dural sac, nerve roots and other posterior appendages, it is almost impossible to insert a large cage through the transforaminal channel [
48]. Additionally, the cages used in OLIF may have a favourable angle between the upper and lower surfaces to induce lordosis (up to 12°), while it is almost plain in MIS-TLIF; therefore, restoration is only achieved by compressing the posterior column [
9,
21,
49‐
51]. Theoretically, OLIF can restore more lordosis and overall spinal alignment than MIS-TLIF. However, our pooled analysis only detected a slight edge with no statistical significance after fusion at L1-5 between the two procedures. We assume that such discrepancy may not be fully reflected in a single-level fusion, and with the increase in fusion segments, the advantages of OLIF in lordosis creation may accumulate and tend to be statistically significant [
27,
51‐
55]. Actually, current studies have partly verified our hypothesis. For example, Champagne et al. reported a mean improvement of 5.5 degrees in lumbar lordosis after multi-level OLIF, far more than a mean improvement of -0.84 degrees after multi-level MIS-TLIF or TLIF [
51]. He et al. and Chen et al. also revealed better improvement of segmental lordosis and overall lumbar lordosis after multi-level OLIF when compared with multi-level TLIF [
27,
52]. However, additional studies focused on MIS-TLIF vs. OLIF are required.
Although pooled analysis found no significant difference between the two methods in terms of fusion rate at the last follow-up, it has been noted that OLIF leads to segmental fusion much more rapidly than MIS-TLIF [
26]. Moreover, subsidence was also significantly lower in the OLIF group. As mentioned above, OLIF provides a much larger annual window to the intended level for comprehensive disc space clearance, endplate preparation and placement of large grafts [
44]. In our experience, the footprint of cages in OLIF is at least twice the size of that in MIS-TLIF and is wide enough to stand on both sides of the dense peripheral apophyseal bone [
56‐
59]. Biomechanical analysis has also suggested that such constructs may effectively diminish stress peaks and disperse the stress in the endplate cancellous bone equally, which is beneficial for both fusion and subsidence resistance [
60‐
62].
The present study had several limitations. First, the level of evidence was relatively low because the studies were all retrospective cohort studies. Second, only single-level fusion was analysed due to a lack of related studies, but we assumed that for multilevel fusion, the advantages of OLIF may be more obvious. Third, important radiographic parameters, such as spinopelvic parameters and sagittal spinal balance, were still missing. Finally, long-term outcome was unclear. Due to the above reasons, high-quality studies are still required to validate the respective advantages of MIS-TLIF and OLIF.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.