Comparison of Early Outcomes between Percutaneous Endoscopic and Minimally Invasive Transforaminal Lumbar Interbody Fusion for Lumbar Spondylolisthesis

Background With the rapid development of less-invasive techniques, the percutaneous endoscopic transforaminal lumbar interbody fusion (Endo-TLIF) as a novel minimal surgical technique for treating lumbar spondylolisthesis in recent years. To compare the preliminary ecacy of Endo-TLIF with that of minimally invasive transforaminal lumbar interbody fusion (MIS-TLIF) for the treatment of lumbar spondylolisthesis.


Background
Lumbar spondylolisthesis is de ned as a forward slippage of a lumbar vertebra relative to the next vertebral body and resulting in instability of the segment [1,2]. The most common types include degenerative and isthmic lumbar spondylolisthesis. Degenerative spondylolisthesis is one of the most common degenerative spinal disorder in the aging population, and often associated with lumbar canal stenosis [3,4]. And, it is also a frequent cause of low back pain, and it results from a narrowing of the disc space or nerve root canal [2]. Bhalla pointed out that the chief goals of surgical treatment include neurologic decompression and stabilization of the vertebral segments with instrumented fusion [5].
Posterior lumbar interbody fusion (PLIF) is regarded as most common surgery for degenerative lumbar disease such as degenerative spondylolisthesis and lumbar spinal stenosis [6]. Although open surgery can have good effects, it may damage the paraspinal muscles, result in postoperative pain, prolong hospital stays, protract rehabilitation programs and increase the nancial burden to patients. Fan et al. reported extensive stripping of muscles, ligament resection, destroy the architecture of the spinal posterior column may result in spinal instability and failed back surgery syndrome (FBSS) [7]. These drawbacks of standard posterior lumbar interbody fusion have prompted the development of lessinvasive techniques [8][9][10][11]. Furthermore, with the improvements in surgical instruments and the increase in patients' demands for quality of life, minimally invasive surgery has been accepted by an increasing number of physicians and patients. Kai-Michael et al [12] pointed out that numerous minimally invasive techniques have been applied for spinal surgery since a novel tubular retractor system was introduced by Foley in 1997. In recent years, a variety of less-invasive procedures that can minimally disrupt normal structures without compromising effectiveness have been applied in the eld of spinal surgery, such as minimally invasive transforaminal lumbar interbody fusion (MIS-TLIF), anterior lumbar interbody fusion (ALIF), extreme lateral lumbar interbody fusion (XLIF) and oblique lumbar interbody fusion (OLIF), each having its own bene ts and drawbacks [13,14].
Harms et al. [15] rst reported transforaminal lumbar interbody fusion (TLIF) technique, which can effectively avoid muscle and nerve root traction injury, in 1982. In 2002, Foley [16] reported that the MIS-TLIF technique, which was derived from TLIF and is performed with a tubular retractor, could effectively preserve back muscles and reduce postoperative complications. In recent years, it has been widely used in the treatment of a variety of lumbar degenerative diseases, with satisfactory outcomes obtained [17].
Many studies have shown that advanced MIS-TLIF is associated with less blood loss and a shorter recovery time than conventional open surgery, but laminotomy, facetectomy and avum dissection are necessary in order to achieve interbody fusion with the use of cages. Meanwhile, this technique still questioned by its limited workspace and the eld of vision of this surgical procedure, steep learning curve, and may higher incidence of complications [18].
Recently, the percutaneous endoscopic transforaminal lumbar interbody fusion (Endo-TLIF) technique was reported to address a variety of spinal disorders by using endoscopic and expandable cages through Kambin's triangle [19,20]. Endo-TLIF technique was derived from the percutaneous endoscopic lumbar discectomy (PELD) technique and combined endoscopic visualization, expandable cage technology, and interbody fusion technique [21]. As reported, the Endo-TLIF technique can achieve not only bilateral direct decompression, interbody cage insertion and pedicle implantation but also less dissection of normal structures. In other words, the muscle, soft tissue and nerve roots can be signi cantly protected because of the access to procedures and the direct visualization under endoscopy. In 2020, AO et al. [22] reported that their study showed that, compared with MIS-TLIF, their percutaneous endoscopic transforaminal lumbar interbody fusion technique has advantages of less surgical trauma, less hidden blood loss, less postoperative low-back pain, and faster recovery. Although both their and our techniques share the label of percutaneous endoscopic transforaminal lumbar interbody fusion technique, they differ in several aspects. The difference not only in the methods of screw implantation, working channel system, surgical procedures and details but also in interbody implant cage. In recent years, research on the Endo-TLIF technique has become a hot topic, and among them, a great many spine surgeons put more attention into the clinical application of the technology, which is an important eld.
There are a limited number of studies that have evaluated the use of the Endo-TLIF technique for the treatment of lumbar spondylolisthesis and compared early clinical outcomes with those of other minimally invasive lumbar fusion surgeries. So that, surgeons lack references related to percutaneous endoscopic lumbar interbody fusion technique and clinical experience in evaluating preliminary clinical outcomes. Thus, the aims of this study were as follow: (i) demonstrate the surgical procedures, technique, advantages and drawbacks of Endo-TLIF and MIS-TLIF; (ii) evaluate and compare the early clinical e cacies of two minimally invasive techniques for single-segment lumbar spondylolisthesis; (iii) provide spine surgeons and patients with accurate early outcome of Endo-TLIF and MIS-TLIF and relevant theoretical basis for the choice of operation method.

Methods
Method approach to clinical study, the Endo-TLIF group compared with the MIS-TLIF group. In total, 62 patients were enrolled in the study between May and August 2019 according to the screening criteria. The procedures were approved by the ethics committee of the A liated Hospital of Qingdao University.
The inclusion criteria were de ned as follows: (i) Patients over 18 years old; (ii) Patients who were diagnosed with degenerative or isthmic spondylolisthesis (Grade I/II) with or without resultant stenosis at one lumbar level; (iii) Patients who were diagnosed with single segment lumbar instability; (iv) Patients who had neural symptoms (including low back pain, sciatica and extremity symptoms); (v) Patients whose symptoms were not alleviated after 6 months or more of conservative therapy.
The exclusion criteria were de ned as follows: (i) Patients with serious systemic diseases or patients who could not withstand surgical treatment; (ii) Patients who had undergone previous lumbar fusion surgery or patients who needed treatment for more than one level; (iii) Patients who were mentally incompetent; (iv) Patients who had trauma, arachnoiditis, active infection (local or systemic) or spinal metastasis.
All diagnoses were based on clinical symptoms and neuroradiological imaging manifestations, especially those of extension and exion lateral radiograph, computerized tomography (CT) and magnetic resonance imaging (MRI) scans. Patients enrolled in the experiment were divided into two groups: the Endo-TLIF group and the MIS-TLIF group. All patients underwent surgery, received general anesthesia in the prone position and were followed up for a period of 12 months. None was lost to follow up in the two groups.

Endo-TLIF Technique
Following induction of general anesthesia, the patients were positioned prone on a radiolucent table.
Spinal cord monitoring was performed during the procedure to prevent unexpected nerve injury. Compared with other Endo-LIF, the instruments called Endo Surgi-Plus® and Endo-TLIF system were designed and manufactured by Unin-tech company from China, and all the procedures were visualized under the endoscopy. The entry point for endoscopic working channel and the insertion of pedicle screws were marked on the skin via uoroscopy. Two or four longitudinal incisions with the length from 1 to 1.5 cm approximately 4-6 cm from the spinous process was made on the side and at the level of pathology based on the entry point. Blunt guiding rod was inserted along the guiding wire of inferior percutaneous pedicle screw and put on the surface of the zygapophyseal joint, followed by working channel. The appropriate position for insertion of the working channel was con rmed with anteroposterior and lateral X-ray views (Fig. 1).
Foraminoplasty and laminectomy were achieved by the removal of part of the superior and inferior articular process with full-see reamers for 4-8 times. A satisfactory foraminoplasty and laminectomy was obtained through multiple reamers cutting and lamina forceps according to preoperative plan for decompression. Caudal, cephalad, lateral edge of ligament avuum were visible clearly and removed with rongeurs. By turning the working channel from a ventral to dorsal position, the nerve root and dural sac were protected behind the sharp bevel of the working cannula. Next, discectomy and nerve root decompression were performed using the Endo Surgi-Plus® and Endo-TLIF system under endoscopic visualization (Fig. 1). Bilateral discectomy and nerve root decompression should be performed for patients with neurological symptoms of both lower limbs through the Endo-TLIF technique.
All the procedures of Discectomy and endplate preparation were nished under endoscopy by using osteotomes, blue forceps, inside and outside reamer, chisel, scraper and exible scraper. Bone graft bed was made into fan shaped, and as good as that in open posterior lumbar interbody fusion (PLIF) surgery, trial cages were implanted with increasing sizes under uoroscopic control (Fig. 1). Based on the results of the trial cage, a suite of expandable interbody cages lled with decompressed skeletal particles was selected and subsequently placed into the disc space through the working tube. The cages were adjusted to a satisfactory position according to intraoperative uoroscopy; then, the cages were expanded to increase the rear height by 2 mm to make full contact with the endplate. The cage was expanded through the special access cannula, then, su cient bony grafts was placed in the expanded cage through a hollow tube (Fig. 2).
Before removing the working channel and prior to endoscopy, the cage position and decompression effect of the nerve root were con rmed again. Four guide wires were placed through cannulated needles that were inserted into the pedicle based on the bilateral pedicle projection. Pedicle screw holes were tapped, and screws and rods were inserted percutaneously (Fig. 2).
Under the guidance of the C-arm X-ray machine, the surgical level was located, and the skin incision was made on the side and at the level of pathology. The muscle (multi dus and longissimus muscles) was bluntly separated from dorsal to ventral, and the quadrant system was placed on the symptomatic side via the muscular gaps. Unilateral laminectomy, facetectomy, partial ligament avum resection, nerve root decompression, excision of the intervertebral disc and a sizeable rigid PEEK cage placement were performed based on the quadrant dilator. The process of pedicle screw and rod placement was the same as that used for Endo-TLIF.

Clinical Outcomes
All data were collected after the acquisition of written informed consent from the patients. The demographic data collected included sex, age, body mass index (BMI), segment distribution and spondylolisthesis severity. Patients were assessed for perioperative factors such as operative time, estimated blood loss (EBL), interoperative uoroscopy time, ambulation time and operative complications. The results of clinical metrics such as the Visual Analog Scale (VAS) for back pain, the Oswestry Disability Index (ODI) and the Japanese Orthopaedic Association (JOA) score were obtained and used to compare the early outcomes between the two groups on preoperative day 1 and on postoperative week 1 as well as months 3, 6 or 12. Postoperative fusion rates were assessed by CT scan 6 to 12 months after surgery. Continuous bone trabecular bridging between intervertebral bodies was considered as the standard of spinal intervertebral fusion [23].

Oswestry disability index (ODI)
Oswestry disability index (ODI) is a principal condition-speci c outcome measures used in the management of spinal disorders, and to assess patient progress in routine clinical practice. If all10 sections are completed the score is calculated as follows: total scored out of total possible score × 100.If one section is missed (or not applicable) the score is calculated: (total score/(5 × number of questions answered)) × 100%.0%-20% is considered mild dysfunction, 21%-40% is moderate dysfunction, 41%-60% is severe dysfunction, and 61%-80% is considered as disability. For cases with score of 81%-100%, either long-term bedridden, or exaggerating the impact of pain on their life.

Japanese Orthopaedic Association (JOA) score
The JOA scoring is one of the most frequently used clinical outcome measures to re ect the improvement of lumbar function before and after treatment. The JOA total score is 29 points and the lowest is 0 points. A higher score indicates more complete the lumbar function.

Data analysis
Statistical analysis was performed using the SPSS 22.0 statistical software package (SPSS Inc., Chicago, IL, USA). The continuous data from the Endo-TLIF and MIS-TLIF groups were assessed with t-tests. Categorical variables were analyzed with the χ2 test. A P-value < 0.05 was considered statistically signi cant.

Perioperative Parameters
The mean age of the 32 (51.6%) patients in the Endo-TLIF group was 53.1 years old (53.1 ± 16), and there were 12 males and 20 females. The mean age of the 30 (48.4%) patients in the MIS-TLIF group was 55.7 years old (55.7 ± 14.2), and there were 14 males and 16 females. No signi cant differences were found in the demographic data of all patients in the two groups. Further details were listed Table 1. Compared with the MIS-TLIF group, the Endo-TLIF group had a lower estimated intraoperative blood loss (73.0 ± 26.4 ml), a shorter ambulation time (1.6 ± 0.6 days) and a higher intraoperative uoroscopy time (46.3 ± 5.1 seconds). In our study, MIS-TLIF need at least 7 times X-rays but Endo-TLIF requires at least 15 times X-rays. A longer mean operation time was observed in the Endo-TLIF group (202.6 ± 31.4 minutes) than in the MIS-TLIF group (192.1 ± 18.9 minutes), but there were no statistically signi cant differences between the two groups. The further perioperative indicators are shown in Table 2.

Functional Evaluation
Both groups showed a signi cant reduction in the VAS scores for back pain and in the ODI score, but the Endo-TLIF group showed a better curative effect at the 6-month follow-up. At more than 6 months postoperatively, both groups showed continuous improvement, but there were no signi cant differences between the two groups. Leg pain VAS scores were similar between the two groups with no signi cant differences in the follow-up period. JOA scores were used to assess neurologic function, and the further results are shown in Fig. 3.

Interbody Fusion
In 2 cases in the Endo-TLIF group, continuous bone trabecular bridging could not be identi ed based on the CT scan at the 12-month follow-up, and the remaining 60 cases in the two groups were regarded as interbody fusion (Fig. 4). However, it must be noted that there were no signi cant differences in the interbody fusion rates between the two groups.

Complications
There were no cases of severe complications, such as postoperative infections, dura mater tears or nerve root injuries. But one case of Endo-TLIF group happened that the shorter rod came off from the screw 7 week later after the surgery and was revised under local anesthesia to change a longer rod.

Discussion
Theoretical Basis and Advantages of Endo-TLIF Endo-TLIF technique was derived from the PELD technique and can be combined with the interbody fusion technique. The present study showed that Endo-TLIF offered many advantages over MIS-TLIF for the treatment of LSS. In our study, the MIS-TLIF technique requires 4-5 cm lengthy skin incision approximately. Although the Wiltse approach with the use of a quadrant channel can reduce the need for muscular stripping, it still requires partial laminotomy to achieve decompression and fusion [18]. Because of the application of endoscopy and progressive tissue dilation, the Endo-TLIF technique is performed with a 1-1.5 cm-length skin incision and entirely preserves the function of paraspinal muscles. Furthermore, because of the smaller tube used, the technique requires a more accurate angle for working channel placement. As has been reported, the limited amount of muscle dissection and innervation damage are principle factors that can lessen the incidence of low back pain and shorten the recovery time [7]. In the Endo-TLIF group, the mean amount of blood loss was 73.0 (± 26.4) ml without postoperative drainage, and the patients were ambulatory within 12 hours. Compared with MIS-TLIF group, continuous irrigation of normal saline and certain water pressure during the operation may be one of the main factors for reducing blood loss.
Conventional fusion is still the widely accepted standard for managing lumbar degeneration disorders that required the maintenance of spinal stabilization. The distinguishing factor of the Endo-TLIF technique is the endoscopy-based transforaminal posterolateral approach that permits the use of an expandable fusion cage placed in the disk space through Kambin's triangle with limited removal of bone structure [24]. Nerve root injury is one of the most serious complications of spinal surgery, so to protect nerve roots, we must pay great attention before each step in the procedure that may result in damage. Foraminoplasty is a necessary step for enlarging the foramen in order to provide su cient space for subsequent procedures and eliminate exiting root injury [19]. Similar to the PELD technique, the nerve root can be protected by a working tube with a beveled tip from trepan cutting and cage implantation. There were no cases of exiting nerve root injury in the study. Endplate preparation and expandable cage placement under direct endoscopic visualization provides an effective advantage for the protection of nerves and the dural sac.
The application of expandable cages is also one of the advantages of the Endo-TLIF technique. Compared with a rigid PEEK cage, this cage requires less excision of bony structures and a smaller size in order to obtain a satisfactory result and to avoid the excessive pulling of nerve roots [25]. Expandable cages are convenient for implantation, and they provide su ciently high interbody restoration during lumbar spine surgery. Spinal leg pain may result from disc and foraminal collapse, which results in compression of the exiting nerve root. In such circumstances, expandable cage implantation can restore intervertebral height to achieve indirect neural decompression and additional foraminal expansion [26]. The Varian® cage (Medyssey, Korea) that we used in Endo-TLIF procedure allowed adjustment of the cage's position once expansion was implemented before the bone graft was placed. In addition, the surface of the expandable cage provides stability and decent bone integration with the help of a bone graft within the cage. Endplate preparation is a vital step for achieving satisfactory fusion. In our study, interbody fusion was obtained in 60 patients and was con rmed with postoperative CT scans, no cage subsidence was observed in two groups. The damage to the endplate in the process of cage implantation might have been the main cause of cage subsidence, so that, satisfactory endplate preparation and careful cage implantation were performed in the process of surgery. In addition, percutaneous pedicle screw xation is necessary in Endo-TLIF, especially for patients who have broken endplates or osteoporosis [27].

Drawbacks of Endo-TLIF
Endo-TLIF technique offers many advantages, such as limited invasiveness, limited intraoperative blood loss, a short recovery time and no need for postoperative drainage. Despite the many bene ts mentioned above, there are some limitations that should be given attention. The locations of the working tube, expandable cage and pedicle screws need to be determined with X-ray during the operation, thus, patients and doctors are exposed to high doses of radiation. It is undeniable that the Endo-TLIF technique still requires a steep learning curve while requiring a thorough understanding of foraminal anatomy and rich endoscopic experience.

Limitation of Study
Some limitations of the study must be noted. First, the number of patients in the two groups was relatively small. Second, all patients in the study were only followed for 1 year, so a larger sample with a longer follow-up time is needed to make de nitive clinical conclusions. Finally, the height of the iliac crest had some in uence on patient selection.

Conclusions
In present study, Endo-TLIF technique has shown great advantages in terms of the early curative effect in patients. As a result of the study, the Endo-TLIF technique seems to be a suitable choice for patients with lumbar spondylolisthesis due to its advantages of being minimally invasive and the fact that it can even be performed under local anesthesia. The use of the Endo-TLIF technique is increasing, and in the near future, it will become a prominent part of spinal surgery.

Availability of data and materials
The data supporting the ndings of the current study are available within the article.

Competing interests
The author reports no con icts of interest in this work