Minimally Invasive versus Conventional Transforaminal Lumbar Interbody Fusion in Treatment of Single-Level Low-Grade Lumbar Spondylolisthesis: A Systematic Review and Meta-Analysis

Background Data: Degenerative lumbar spine, including spondylolisthesis, is a common clinical condition that affects humans in the most productive period of their life. There are many surgical options for the management of such conditions after the failure of conservative therapy. Recently, there has been a great debate regarding the use of minimally invasive (MI) versus open transforaminal lumbar interbody fusion (O-TLIF) in the treatment of single-level low-grade lumbar spondylolisthesis, so there was a need to reach a consensus over this issue. Purpose: To compare the clinical efficacy and safety of MI-TLIF versus O-TLIF in the treatment of single-level low-grade degenerative lumbar spondylolisthesis. Study Design: A systematic review for recent studies in the context and meta-analysis. Patients and Methods: We searched online databases of PubMed, Google Scholar, Cochrane Library, and DOAJ (2016–2020), and the search yielded 1352 articles. Based on our inclusion and exclusion criteria, we included retrospective, prospective, and randomized control trials, which came down to 11 research articles. Operative time, blood loss, hospital stay, back pain scores (Visual Analogue Scale), functional score (Oswestry Disability Index), complication rate, and reoperation rate for both techniques were recorded and presented as means. We then performed a meta-analysis. Results: There is an overall advantage for the MI-TLIF over the O-TLIF in different parameters. There was a statistically significant difference in blood loss of −0.954 ml (p = 0.000) and hospital stay of −1.19 days (P = 0.000), favoring M-TLIF. There was a statistically insignificant difference in the total operative time (P = 0.071), the postoperative VAS of −0.22 (P = 0.384), and the postoperative ODI of −2 (P = 0.331). Moreover, there was a reduced combined odds ratio for complications and a reduced odds ratio for re-operation. SYSTEMATIC REVIEW EgySpineJ 39:2-17, 2021


INTRODUCTION
Degenerative spondylolisthesis (DS) is an acquired anterior-vertebral displacement without disrupting the pars interarticularis, associated with the degenerative changes of aging, such as intervertebral disc degeneration, ligamentous hypertrophy or buckling, and osteophyte proliferation. 13,25,1 This clinical condition place enormous socioeconomic and health burdens on the health service providers and society. Instrumented lumbar interbody fusion (LIF) is a commonly used surgical intervention to treat various kinds of lumbar disease requiring fusion. Recently, LIF using minimally invasive techniques, such as percutaneous pedicle screw fixation (PPSF), has been used frequently with the advancement of minimally invasive spinal technique (MIS). 5,10,17,36,38 The preferred approaches for this procedure are posterior lumbar interbody fusion (PLIF) 30,31 or transforaminal lumbar interbody fusion (TLIF). 3,7,11,21,29 In 2002, Foley and Lefkowitz 6 first introduced the minimally invasive transforaminal lumbar interbody fusion (MI-TLIF) technique. With the advancement of surgical instrumentation and optical systems, the MIS-TLIF technique has become more and more popular with the potential advantages of smaller wound size, less tissue trauma, and faster recovery. 27,35,37 Recently, other approaches 12,47 have been performed; however, MI-TLIF has gained more popularity than others due to no thecal sac retraction and the lower level of trauma to back muscle and bony structures such as facet joints and lamina. Although many articles have reported O-TLIF or MI-TLIF, no studies have reported the long-term clinical and radiological outcomes of instrumented MI-TLIF. Other studies have reported the harmful effects of extensive muscle dissection and excessive blood loss due to this traditional O-TLIF procedure. 22,34,39 Up to now, no consensus has been reached regarding which procedure can achieve better effects in the treatment of symptomatic lumbar spondylolisthesis. 35 This study was performed to estimate the clinical efficacy and safety of MI-TLIF versus O-TLIF in the management of single-level low-grade degenerative lumbar spondylolisthesis.

PATIENTS AND METHODS
Search Strategy: This study was performed according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA). 23 The relevant literature retrieval was performed in 4 electronic databases, including PubMed, Google Scholar, Cochrane Library, and Directory of Open Access Journal (DOAJ). The final searches were performed on January 5 th , 2021. Reference lists of included articles and relevant meta-analysis were manually searched. Randomized or nonrandomized controlled studies published from January 2016 to December 2020 that compared MI-TLIF with O-TLIF for the treatment of low-grade lumbar degenerative spondylolisthesis were retrieved. We searched these databases using a combination of the keywords and medical subject headings. For maximum sensitivity of the search strategy, the search terms were combined as follows: 1) transforaminal lumbar interbody fusion OR TLIF OR open; 2) minimally invasive transforaminal lumbar interbody fusion OR MI-TLIF OR minimally invasive surgery; 3) single-level degenerative spondylolisthesis; 4) 1, 2, and 3. Only articles that were published in the English language were included. Citations abstracts and full manuscripts were downloaded and de-duplicated for screening and categorization of potentially eligible studies. For degenerative spondylolisthesis, the initial searches were conducted independently by two reviewers (MHM, MS) to screen all retrieved titles and abstracts. Unqualifying studies were initially excluded, while the full text of eligible reports was assessed. The reference lists of all acquired articles were also manually checked for additional relevant studies. Discrepancies between them were resolved by discussion. Inclusion Criteria. Eligibility criteria for study selection included in the present network metaanalysis are as follows: (1) an RCT and non-RCT published in English; (2) patients with degenerative lumbar spondylolisthesis; (3) comparing the 2-fusion procedure, MIS-TLIF, and O-TLIF; (4) treatment-specific outcomes including preoperative and postoperative VAS (Visual Analogue Score) and Oswestry Disability Index (ODI) scores, blood loss, operative time, hospital stay, reoperation rates, and complications; (5) an average follow-up duration of at least 12 months. Exclusion Criteria: Studies were expelled according to the following items: (1) <10 patients per intervention arm, 19 (2) Observational studies, case reports, conference abstracts or paper, and duplicated papers or reviews, and (3) Qualified data from the original studies could not be extracted.

Search Results:
We searched online databases of PubMed, Google Scholar, Cochrane Library, and DOAJ (2016-2020), which yielded 1352 articles. We included retrospective, prospective, and randomized control trials based on our inclusion and exclusion criteria, which came down to 11 research articles. A PRISMA flowchart diagram depicting the study identification and selection process is shown in Figure 1. Data were extracted independently and duplicated from eligible studies by the same two researchers using standardized data collections forms developed a priori. Data items recorded included general manuscript information, patients' characteristics, study characteristics, treatment details, and main outcomes (Table 1). Data extraction discrepancies between the two researchers were resolved by discussion. Moreover, we have applied the quality check on the papers included according to the 8-Item Modified Jadad Scale, as explained in Table 2. Operative time, blood loss, hospital stay, pain scores (Visual Analogue Scale), functional score (Oswestry Disability Index), complication rate, and reoperation rate for both techniques were recorded and presented as means. We then performed a meta-analysis.

RESULTS
Eleven studies were reported in this systematic review, including four randomized controlled trials (RCT) 40,43,46,48 and seven nonrandomized controlled trials. 2,18,24,28,33,42,44 The summary of our extracted data and reported articles is presented in Based on our meta-analysis, there was no statistical significance between both procedures (P ≤ 0.071) ( Figure 2).

Blood Loss:
Ten studies had sufficient data regarding the amount of operative blood loss. The mean operative blood loss volume was 149.13 ± 77.26 ml in the MI-TLIF group and 287.44 ± 127.12 ml in the O-TLIF group. The difference was significant and favored the MI-TLIF procedure (P ≤ 0.001) ( Figure 3).

Hospital Stay:
Eight studies had sufficient information on the length of hospital stay. The mean hospital stay was 5.3 ± 2.9 days in the MI-TLIF group and 7.12 ± 3.9 days in the O-TLIF group. The difference was significant and favored the MI-TLIF procedure (P ≤ 0.001) (Figure 4).

Low Back Pain Visual Analogue Score:
Six studies had sufficient data regarding the VAS scores of LBP. The mean preoperative VAS score for LBP was 6.45 in the MI-TLIF group and 6.37 in the O-TLIF group, with no statistically significant difference (P = 0.388) ( Figure 5). The mean VAS score for postoperative LBP at the final follow-up was 1.19 in the MI-TLIF group and 1.41 in the O-TLIF group with no statistically significant difference between both procedures (P = 0.137) ( Figure 6). There were marked differences and significant improvement between the preoperative and the postoperative VAS at the final follow-up in both procedures.

Oswestry Disability Index:
Six studies reported sufficient data on the ODI scores expressed in percentage. The mean preoperative ODI score was 46.38 in the MI-TLIF group and 45.13 in the O-TLIF group. The difference between both groups was not statistically significant (P = 0.320) (Figure 7). At the final follow-up, the mean ODI score was 18.63 in the MI-TLIF group and 20.63 in the O-TLIF group, with no significant difference between both groups (P = 0.331) ( Figure 8). There were marked differences and significant improvement between the preoperative and the postoperative ODI at the final follow-up in both procedures.

Complications:
The number and details of complications have been reported in seven studies. The complication rate was 2.14% in the MI-TLIF group and 2.28% in the O-TLIF group. The difference between both groups was not statistically significant (P = 0.634) ( Figure 9). Reported complications in seven studies were minor in general and included incidental dural tear, added neurological deficit, screw malposition, cage migration, wound infections, delayed wound healing, pseudoarthrosis, large seroma, large symptomatic seroma, contralateral radiculopathy, myocardial infarction, urinary tract infections, and bowel and bladder incontinence.

Reoperation Rate:
Four studies reported sufficient data on the reoperation rate expressed in percentage. The mean percentage of reoperation in the MI-TLIF group was 2% and 6% in the O-TLIF group, without any statistically significant difference between the two groups (P = 0.758) ( Figure 10). The most common causes of reported reoperation in the study articles were adjacent segment disease, pseudoarthrosis, surgical site infection, contralateral radiculopathy, and implant-related complications, including cage and screw repositioning.         6,14,15 For beginners, the challenges of MI-TLIF lie in the steep learning curve and the longer operative time. 18 We reviewed previous similar systematic reviews reporting the outcome of MI-TLIF and O-TLIF in treating single-level low-grade lumbar spondylolisthesis or mixed indications and reported three studies. 8 20 Mummaneni et al. 28 found no difference with regard to the length of hospital stay and 90-day return-to-work period. Su et al.'s 42 concluded in their study that in low-grade degenerative spondylolisthesis, both MI-TLIF and O-TLIF were associated with a significant reduction in vertebral slip; however, O-TLIF had a higher rate of slip reduction than MI-TLIF. They also reported that MI-TLIF significantly reduces lumbar lordosis and slip angle, resulting in relative kyphosis at the fused segment. Finally, they found that O-TLIF significantly reduces L1 axis and S1 distance and may be more conducive to improving lumbar sagittal balance. Contrary to their results, Serban et al. 40 reported similar radiological outcomes parameters among both M-TLIF and O-TLIF surgical groups.  35 Hammad et al. 8 Miller et al. 26 Kim et al. 16 Search This review has some limitations, including the paucity of RCTs and some reported studies not documenting the radiation exposure, fusion, sagittal balance, opioids use, and perioperative cost of each procedure. Multilevel and highgrade spondylolisthesis patients not reported here warrant more studies. More RCTs with a long-term follow-up are highly recommended with a focus on items mentioned in the limitations. Furthermore, important limitations are that some papers mixed other diagnoses with spondylolisthesis in the count pool of subjects, while some other papers counted grades I and II in the same counting pool.

CONCLUSION
The reported data in this systematic review and meta-analysis suggest that there was a significant difference in operative blood loss, and hospital stay between both groups that favor MI-TLIF versus O-TLIF procedure. While there was no significant difference in operative time, VAS, ODI, reoperation rate, and rate of postoperative complications between both groups.