An Innovative Percutaneous Endoscopic Transforaminal Lumbar Interbody Fusion for the Treatment of Lumbar Spinal Stenosis with Degenerative Instability: a prospective cohort study

Lumbar spinal stenosis (LSS) is most common lumbar degenerative diseases for people with low back pain. Endoscopic lumbar fusion technique was considered as a promising treatment for LSS with degenerative instability. The objective of this study was to compared the clinical effects for the treatment of Lumbar spinal stenosis (LSS) with degenerative instability between the innovative percutaneous endoscopic transforaminal lumbar interbody fusion (PE-TLIF) technique and posterior lumbar interbody fusion (PLIF) technique. Between April 2019 and December 2019, 40 patients with single-segment LSS were prospectively included in our study. Visual Analogue Scale (VAS) on lumbar and leg pain (VAS-LBP, VAS-LP), Oswestry Disability Index (ODI), serum Creatine Kinase (CK), the maximal cross-sectional area of multi�dus muscle (Max-CSA) and the peak intensity of Sulphur hexa�uoride microbubble contrast agent (PI) around the surgical incision by contrast-enhanced ultrasonography were evaluated preoperatively, post-operatively and at regular follow-up. All patients were followed up. The VAS-LBP, VAS-LP, ODI after operation were improved signi�cantly compared to these data before operation in all the patients (P < 0.05). The VAS-LBP at 1 weeks, 3 months after operation in PE-TLIF group were signi�cantly lower than these in PLIF group (P < 0.05). The injury degree of multi�dus muscle evaluated by MAX-CSA and PI was signi�cantly less in PE-TLIF group after operation(P < 0.05). There was no signi�cant difference on the complication rate between these two groups (P > 0.05). Our results presented PE-TLIF technique could obtain comparable effective outcomes as conventional PLIF for the treatment of LSS with degenerative instability. The Patients with PE-TLIF had less muscle injury, less pain and quicker postoperative rehabilitation. (A multicenter non-randomized controlled study for percutaneous endoscope transforaminal lumbar interbody fusion (PE-TLIF) and traditional open surgery for the treatment of degenerative diseases of the lumbar spine;2019/4/13; http://www.chictr.org.cn/showproj.aspx? proj=38002; ChiCTR1900022492)


Introduction
It is reported that low back pain is the leading cause of elder people lived with disability from 1990 to 2017 [1], and lumbar spinal stenosis (LSS) is most common lumbar degenerative diseases for people with low back pain [2].Although most of patients with LSS could be treated effectively by conservative method, there still are some patients require surgical treatment due to the failure to alleviate the severe pain by conservative treatment [3].
Spine fusion surgery has been considered as an effective method in improving pain, segment stability, function and quality of life in patients with LSS, especially the patients with degenerative instability [4,5].
Most of surgeons consider posterior lumbar interbody fusion (PLIF) as the standard operation for LSS with degenerative instability, and the PLIF demonstrated a satisfactory clinical effect and a higher fusion rate [4].Nevertheless, extensive destruction of posterior muscular-ligamentous complex usually leads to muscular atrophy, tremendous postoperative pain, and functional disability [6,7].Therefore, minimally invasive surgeries gradually gained popularity in order to overcome the disadvantages of traditional open surgeries in the past twenty years.
Recently, endoscopic lumbar fusion techniques have been attempted successfully in the treatment of patients with LSS [8][9][10][11], however, some drawbacks still exist, including nerve root injury, cage-related complication, long learning curve and so on.To overcome the aforementioned disadvantages, we developed an innovative percutaneous endoscopic transforaminal lumbar interbody fusion (PE-TLIF) with oriented superior articular process (SAP) resection device, and showed a good clinical result in a preliminary report [12].As best we know, all the studies on endoscopic lumbar fusion technique for the treatment of patients with LSS were retrospective, so the clinical evidence is relative lower.Hence, we conducted a prospective cohort study on PE-TLIF for the treatment of patients with LSS in order to provide a high-level evidence for clinical practice.Meanwhile, we compared the injury degree of lumbar multi dus muscle between PLIF and PE-TLIF for the treatment of LSS with degenerative instability.We hypothesized that the clinical effects on PE-TLIF for the treatment of LSS with degenerative instability was not inferior than PLIF.

Materials And Methods
Between April 2019 and December 2019, 40 patients diagnosed as single-segment LSS with degenerative instability were prospectively included in our study (ChiCTR1900022492).The eligible criteria were: (1) patients were treated by PLIF or PE-TLIF; (2) no history of lumbar surgery; (3) no obvious multi dus muscle injury certi ed by ultrasound.(4) no lumbar scoliosis or deformity.Our study was approved by the institutional review board of Beijing Chaoyang Hospital.All patients were told all possible results on these two surgeries and signed written consent before operation.
Appropriate clinical and radiological assessments were performed for all patients before a decision of operation.18 patients were operated by PE-TLIF technique.There were 4 males and 14 females, with an average of 60.50±9.56years.22patients were operated by PLIF technique.There were 6 males and 16 females, with an average of 60.64±7.42years.Operative level was L4/5.The Bridwell criteria was used to evaluate the intervertebral fusion via CT at 6 months after operation.Visual Analogue Scale (VAS) on lumbar and leg pain (VAS-LBP, VAS-LP), Oswestry Disability Index (ODI), serum Creatine Kinase (CK) were evaluated preoperatively, post-operatively and at regular follow-up.We also employed contrast-enhanced ultrasonography to calculate the maximal cross-sectional area of multi dus muscle (Max-CSA) and the peak intensity of Sulphur hexa uoride microbubble contrast agent (PI) around the surgical incision to present the multi dus muscle condition (Figure 1).More demographic characteristics data of the two groups are listed in Table 1.There was no statistical signi cance between the two groups before the operation in age, sex distribution, operative level, VAS-LBP, VAS-LP, ODI, CK (U/L), Max-CSA, and PI (Table 1).

PE-TLIF
The patient was in a prone position under general anesthesia or low dose epidural anesthesia combining with local anesthesia.The aimed lumbar segment was con rmed under the C-arm uoroscope.The primary guide pin was inserted into pedicle on the symptomatic side under the guidance of uoroscope, and the secondary guide pin was positioned at SAP through a specially designed SAP guider (Figure 2).Then dilating cannulas were inserted progressively via the secondary guide pin.A hook-shaped front of the cannula was employed to be a protection cannula in order to excise the majority of SAP safely by trepan (Figure 3).The working channel was placed through Kambin's triangle, and the endoscope system was connected.The complete endplate preparation was performed after the canal and nerve root were decompressed, and then an expandable cage was then inserted through the working tube after iliac bone autograft was implanted (Figure 4-5).The spinal canal was examined via endoscope to con rm the nerve root was totally relieved.4 pedicle screws and 2 rods were inserted percutaneously.A drainage tube was placed in the decompression working channel, and the incisions were sutured.More details on PE-TLIF technique were reported in our previous study [12] (Supplemental Video).

PLIF
The patient was in a prone position under general anesthesia.The symptomatic segment was con rmed by the C-arm uoroscope.The posterior middle approach was performed, and subperiosteal stripping was adopted until reaching bilateral facet joints.Bilateral pedicle screws were inserted and then interlaminar fenestration was performed bilaterally.The complete endplate preparation was performed after the nerve root were decompressed.Then a proper cage was inserted after intervertebral bone grafting.The loosening of the nerve root was inspected again, and a drainage tube was placed, and nally the incision was sutured.

Post-operative protocol
The operation time, intraoperative bleeding volume, Incision length, postoperative drainage volume, postoperative bedridden time, and complications were recorded.VAS, ODI, CK, Max-CSA, and PI were evaluated preoperatively, post-operatively and at regular follow-up.The data were analyzed by SPSS 17.0 software with chi-square and Fisher's exact test in nominal data and independent t-test in continuous data.

Results
The mean operation time was 204.17±47.90minutes in PE-TLIF group, and 99.77±30.02minutes in PLIF group.The average of postoperative drainage volume was 41.94±28.65 ml in PE-TLIF group, and 285.23±142.17ml in PLIF group.The mean postoperative bedridden time was 23.11±6.15hours in PE-TLIF group, and 51.64±13.65 hours in PLIF group.The average of intraoperative bleeding volume was 105.56±76.79ml in PE-TLIF group, and 241.82±129.64 ml in PLIF group.The mean incision length was 8.44±2.15cm in PE-TLIF group, and 10.50±1.85cm in PLIF group.There was signi cant difference in the operation time, postoperative drainage volume, postoperative bedridden time, intraoperative bleeding volume and the incision length between the two groups (P<0.05,Table 2).There was no signi cant difference on the intervertebral fusion rate at 6 months after operation between these two groups (P>0.05,Table 2).More details were listed in Table 2.
All patients were followed up, and the average of follow-up period was 15.33±3.07months in PE-TLIF group, and 15.82±2.95months in PLIF group.The VAS-LBP and VAS-LP at 1 week, 3 months, 6months after operation and at nal follow-up were improved signi cantly compared to these data before operation in all the patients (P<0.05,Table 3).The ODI at 3 months, 6months after operation and at nal follow-up were improved signi cantly compared to the data before operation in all patients (P<0.05,Table 3).There was signi cant difference on VAS-LBP at 1 weeks, 3 months after operation between these two groups (P<0.05,Table 3).There was no signi cant difference on VAS-LP and ODI after operation between these two groups (P>0.05,Table 3).More details were listed in Table 3. (Figure 6) The CK at 1 day after operation was higher compared to this item before operation in these two groups (P<0.05,Table 4), and there was no signi cant difference on CK at 1 week after operation in PE-TLIF group compared with this before operation (P>0.05,Table 4), but the CK at 1week after operation still higher in PLIF group(P<0.05,Table 4).The CK at 1 days, 1 weeks after operation in PE-TLIF group was signi cantly lower than these in PLIF group (P<0.05,Table 4).The Max-CSA at 1 week after operation was higher compared to this item before operation in these two groups (P<0.05,Table 4), and there was no signi cant difference on Max-CSA at 3 months, 6 months, and nal follow-up after operation in PE-TLIF group compared with these before operation (P>0.05,Table 4).The Max-CSA at 1 week after operation in PE-TLIF group was signi cantly lower that this in PLIF group (P<0.05,Table 4), but the Max-CSA at 3 months, 6 months and nal follow-up after operation in PE-TLIF group was signi cantly higher that these in PLIF group (P<0.05,Table 4).The PI at 1 week after operation was higher compared to this item before operation in these two groups (P<0.05,Table 4), and there was no signi cant difference on PI at 3 months, 6 months and nal follow-up after operation in PE-TLIF group compared with these before operation (P>0.05,Table 4).The PI at 3 months, 6 months and nal follow-up after operation in PE-TLIF group was signi cantly higher that these in PLIF group (P<0.05,Table 4).More details were listed in Table 4.
There was no signi cant difference on the complication rate between these two groups (P>0.05,Table 2).In PE-TLIF group, 1 patient suffered temporary knee tendon hyperre exia after surgery, and recovered within 24 hours after surgery.In PLIF group, 2 patients experienced incision infection, and were cured by intravenous antibiotics.

Discussion
This was the rst prospective cohort study on the clinical effects between lumbar endoscopic fusion surgery and PLIF for the treatment of LSS with degenerative instability at present.Our present results showed that the clinical effects on PE-TLIF for the treatment of LSS with degenerative instability was not inferior than PLIF, and the muscle injury degree in PE-TLIF technique was superior to PLIF operation, and the postoperative drainage volume, postoperative bedridden time, and postoperative VAS-LBP was signi cantly better for the patients treated by PE-TLIF surgery.
PLIF demonstrated a satisfactory clinical effect for the treatment of lumbar spinal stenosis [4,13,14].
Nevertheless, extensive destruction of posterior muscular-ligamentous complex usually leads to muscular atrophy, tremendous postoperative pain, and functional disability [6,7].Some researchers found that bilateral stripping of the multi dus on PLIF was related to paraspinal muscle atrophy, and about 20% patients with failed back surgery syndrome was associated with paraspinal muscle atrophy [15,16].Kalichman et al found that there was a causal relationship between changes in the paraspinal muscles and low back pain, and a higher density of paraspinal muscles could decrease the symptom of low back pain [17].In addition, Ranger et al showed that the extent of paraspinal muscle atrophy was associated with postoperative low back pain [18].In addition, Khan et al described that back-muscle morphometry should be included as a predictor of clinical outcomes in order to improve postoperative functional results and reduce the surgery-related complications [19].Hence, more and more surgeons believed that decreasing the injury extent of paraspinal muscles was very important to improve the postoperative functional outcomes and reduce the surgical complication rate.

Minimally invasive spine surgeries (MISS) gained popularity to overcome the drawbacks on traditional
open surgeries.There are some advantages on MISS, including minimal soft tissue injury, satisfactory clinical effects, reduced the occurrence rate of surgical complication, and better cost-effectiveness [20].[21].

Schwender et al rstly introduced the minimally invasive transforaminal lumbar interbody fusion (MIS-TLIF), and MIS-TLIF technique presented the potential advantages over traditional open techniques
Then, the MIS-TLIF became popular, and obtained satisfactory clinical improvement and fusion rate [22,23].Although the MIS-TLIF could minimize the injury to normal anatomic structures, the technique still requires an open incision of the posterior muscular-ligamentous complex for tube placement.Therefore, some surgeons have attempted endoscopic lumbar fusion techniques.Said et al. rstly reported endoscopic fusion techniques for the treatment of lumbar degenerative diseases.The technique could acquire good clinical outcomes, but the complication rate was relative higher [10].Then Jacquot et al did not recommend the kind of technique because of the 36% complication rate [11].Hence, decisive technical improvements have been made to reduce the complication rate, including cage improvement, approach modi cation, and intraoperative visualization enhancement.We also developed our technique named PE-TLIF, and initial clinical results was favorable [12].
In our study, we compared the effectiveness of PE-TLIF and PLIF for the treatment of LSS with degenerative instability.PE-TLIF could obtained similar clinical effects and less muscle injury degree.CK, Max-CSA, and PI were employed to investigate the muscle injury in our study.In the follow-up after operation, these muscle injury related indicators was signi cantly better in PE-TLIF group than these in PLIF group.The postoperative rehabilitation was signi cantly improved in PE-TLIF, such as postoperative bedridden time and postoperative VAS-LBP.The compilation rate was low, and only 1 patient suffered temporary knee tendon hyperre exia.Nerve root injury and cage migration was not included in our study.Our technique made decisive technical improvements, including innovative expandable cage and modi ed approach.The learning curve on our technique is not very steep, and the technique is easy to be popularized.
The major advantage of our technique is that we innovatively develop some instruments such as guided SAP resection device and parallel expandable cage.We also improved the diameter of working channel to protect the exiting nerve root and the traversing nerve root, and to the bene t of cage insertion via percutaneous surgery.In our clinical experience, the standard operation procedure of guided SAP resection device is very important on our technique, and the SAP device has the ability to ensure the SAP su cient and safe resection.We also used the innovative hook-shaped front of the cannula to restrict the depth of trepan-cutting SAP to protect the exiting nerve root and dura mater.Endplate preparation played an important role in the fusion aspect, the appearance of hemorrhagic exudation from bone endplate was acceptable under the endoscopic visualization.We also recommend iliac bone autograft and adequate bone graft (≥5mm 3 per intervertebral space).
To the best of our knowledge, this is the rst prospective cohort study on percutaneous endoscopic transforaminal lumbar interbody fusion for the treatment of LSS with degenerative instability.The muscle injury evaluation between endoscopic lumbar fusion surgery and traditional open surgery was rstly reported in our study.All surgeries were performed by one senior surgeon.A number of data on the characteristics of patients, treatment results and complications were reported in our study.However, certain limitations need be addressed.The number of patients is relatively small.More prospective randomized controlled trials are needed to overcome the limitations of our study.

Conclusion
Our results presented PE-TLIF technique could obtain comparable effective outcomes as conventional PLIF for the treatment of LSS with degenerative instability.The Patients with PE-TLIF had less muscle injury, less pain and quicker postoperative rehabilitation.

Declarations Ethics approval and consent to participate
Our study was approved by the institutional review board of Beijing Chaoyang Hospital.All patients were told all possible results on these two surgeries and signed written consent before operation.

Consent for publication
Not applicable.

Availability of data and materials
These were included in our manuscript.Tables Contrast-enhanced ultrasonography demonstrates the blood perfusion of the multi dus muscle microcirculation.
Figure 2 Fluoroscopic insertion of guide pins.A, the primary guide pin (left), the front end of which is threaded design, which can be rmly xed in the pedicle, and the position of the primary guide pin is easily recognized under uoroscopy; the primary guide pin is percutaneously inserted into the vertebral pedicle, rotating to x (right).B-C, C-arm anteroposterior and lateral uoroscopy con rms that the primary guide pin enters the pedicle, and the upper edge of the thread is lower than the dorsal lateral level of the superior articular process.(image from the other patient) D, Physical view of the specially designed SAP guider, the rst guide pin and the second guide pin are connected by the connecting arch, and the second guide pin puncture angle and depth can be adjusted on the connection arch.E-F, C-arm anteroposterior and lateral uoroscopy con rms that the second guide pin is xed to the posterior aspect of the superior articular process.
Resection method of the superior articular process.The hook-shaped protective sleeve clings to the lateral periosteum of the superior articular process, reaches the ventral side of the articular process, protects the exiting nerve root and can control the cutting depth of the trephine at the same time, protects the dura mater and nerve root, and rotates the trephine to remove the superior articular process.
Bone graft bed preparation.The width adjustable reamer and endplate curette are used to prepare the cartilage endplate to adequately expose the bony endplate.Finally, Intervertebral space is fully prepared and the appearance of exudation from bone endplate is good, the bony endplate is fully exposed.

Table 1
Comparison of the demographic characteristics data between two groups