Comparative Evaluation of Posterior Percutaneous Endoscopic Cervical Discectomy Using 3.7mm Endoscopic and 6.9mm Endoscopic for Cervical Disc Herniation: a retrospective comparative cohort study.

Background. Posterior percutaneous endoscopic cervical discectomy (p-PECD) is an effective strategy for cervical diseases which working cannula ranges from 3.7 mm to 6.9 mm. However, no studies were performed to compare the clinical outcomes of endoscopes with different diameters in cervical disc herniation (CDH) patients. The purpose of this study was to compare the clinical outcomes of unilateral CDH patients treated with p-PECD applying the 3.7mm endoscopic with those treated with the 6.9mm endoscopic. Methods. From January 2016 to June 2018, totally 28 consecutive patients presented with single-level CDH who received p-PECD using the 3.7mm endoscopic or the 6.9mm endoscopic were enrolled. The indications for this study were as follows: (1) Unilateral cervical spondylotic radiculopathy with pain irradiated to upper extremity, (2) MRI and CT scan show that foraminal CDH located lateral to the edge of spinal cord, from C4–C5 to C7–T1, (3) Unilateral symptoms caused by foraminal stenosis, (4) Failure after conservative treatment for at least 6 weeks or neurological symptoms aggravated. Patients were evaluated neurologically pre- and postoperatively. The clinical outcomes, including the operation time, the hospitalization, the visual analogue scale (VAS) and the modied MacNab criteria, were evaluated. Cervical uoroscopy, CT, and MRI were performed during follow up. 76.5 minutes compared 61.5

endoscopies were signi cantly improved compared with that before operation (p 0.05).
Conclusion. Both 3.7mm endoscopic and 6.9mm endoscopic are effective methods for CDH in selected patients, and there is no signi cant difference in clinical outcomes. 6.9mm endoscopic is superior to 3.7mm endoscopic in the e ciency of "V" point identi cation, overlying soft tissue removal and spinal cord injury prevention. Furthermore, 6.9mm endoscopic is inferior to 3.7mm endoscopic in anterior decompression of the intervertebral foramen.
Subsequently, in order to minimize the surgical complications of ACDF, various surgical techniques have been carried out. Recently, PECD for the treatment of spinal diseases has become favorable. It offers the advantage of reducing trauma, promoting faster rehabilitation (9)(10)(11)(12)(13)(14)(15)(16)(17)(18)(19)(20)(21) and has a similar short-term clinical bene t to ACDF (22). PECD can be performed through the anterior approach and posterior approach (21) which depends on the site of pathology. Studies reported that anterior percutaneous endoscopic cervical discectomy (a-PECD) possesses the disadvantage of higher probability of intervertebral space decrease postoperatively due to the violation of the intervertebral disc (23). However, p-PECD, as a less invasive techniques with potential advantages, does not have this shortcoming.
However, no studies were performed to compare the clinical outcomes of the 3.7mm endoscopic and 6.9mm endoscopic in CDH patients. Thus, the purpose of this study was to compare the clinical outcomes of unilateral CDH patients treated with p-PECD applying the 3.7mm endoscopic with those treated with the 6.9mm endoscopic.

Patient Characteristics
In the retrospective study we enrolled 28 patients with CDH who underwent p-PECD with 3.7mm endoscopic or 6.9mm endoscopic. All the procedures were performed by the same surgeon from June 2016 to July 2018. Besides, the demographic characteristics of the 28 patients in two groups were also collected.

Inclusion Criteria
The indications for p-PECD were as follows: (1) Unilateral cervical spondylotic radiculopathy with pain irradiated to upper extremity, (2) MRI and CT scan show that foraminal CDH located lateral to the edge of spinal cord, from C4-C5 to C7-T1, (3) Unilateral symptoms caused by foraminal stenosis, (4) Failure after conservative treatment for at least 6 weeks or neurological symptoms aggravated (21,29,30).

Endoscopic Instruments
The details were described in table 2. 3.7mm endoscopic: After general anesthesia, the patient was placed in a prone position. Then intraoperative neurological monitoring (INM) was performed by a surgeon, specializing in neurosurgery.
The surgeon and assistant stranded on the same side of the pathology, and the endoscopic monitor was placed on the opposite side of the pathology ( Figure 1). Kerrison punch and endoscopic drill were applied to perform the laminoforaminotomy or foraminotomy. An endoscope characterized by an inner diameter of 3.7mm and 30° optics angle was also utilized (Shanghai Maoyu medical equipment CO., LTD. China).
Besides, all manipulation was carried out under continuous irrigation of saline solution. The lamino-facet junction should be observed on the radiographic of true antero-posterior view in order to identify the entrypoint. A Kirschner needle, with 18-gauge and 25cm, was inserted and placed on the level of pathology. A 1cm surper cial skin incision was done. Then, the obturator was installed and applied to feel the "V" point which is an anatomcal landmark con uenced by superior border of the inferior laminae, inferior border of the upper laminae, and the medial point of the facet joint ( Figure 1). Hence, the working cannula was advanced and obturator was removed ( Figure 2). At this point, the endoscope was inserted through the working cannula. Radiofrequency probe (joimax ® GmbH, Germany) and endoscopic forceps was used to coagulate and remove the overlying soft tissue, under continuous irrigation with normal saline. Once the osseous anatomical structure has been observed, the inferior border of the upper laminae was resected with endoscopic drill until the ligamentum avum was exposed, and then the endoscopic drill was directed caudally toward the cervical pedicle and laterally toward the facet joint. Finally, in order to expose the exiting nerve root, the ligamentum avum and foraminal ligament were removed.
Subsequently, the underlying disc space was detected with a dissector. Intraoperatively, to prevent excessive removal of the facet joint, a nerve hook was applied to feel the medial wall of the pedicle. After the exiting nerve root exposed successful, the intervertebral disc can be detected. The herniated cervical disc was removed through the shoulder or axilla of the exiting nerve root according to the lesion location ( Figure 3). It is critical to palpate the exiting nerve root using a nerve hook and should feel free after intervertebral disc adequately removed (30,31). Figures 4-6 show the case of a 60-year-old male with cervical discomfort with radiation pain of right shoulder and upper extremity. The patient was diagnosed with a C6-C7 CDH. Postoperative satis ed clinical results were achieved in the patient.
6.9mm endoscopic: The surgical tools applied in this technique were a little different from those applied in 3.7mm endoscopic. A larger endoscope with 6.9mm inner diameter (Figure 7) was used in this approach (Shanghai Maoyu medical equipment CO., LTD. China). The manipulation of foraminotomy, laminoforaminotomy and discectomy ( Figure 8) was the same as 3.7mm endoscopic. A 54-year-old male patient presented with neck pain and right upper extremity numbness. He was diagnosed with a C5-C6 right foraminal CDH (Figures 9-11). Postoperative good clinical results were obtained in the patient.

Statistical Analysis
The 2-sample t test, Wilcoxon signed rank test and the Mann-Whitney U test were applied to compare parametric data between the 3.7mm endoscopic and the 6.9mm endoscopic. P<0.05 was regarded as the threshold of signi cance.

Patients and surgical characteristics
The demographic characteristics of the 28 patients in two groups are listed in table 2.  (Figure 12) between 3.7mm endoscopic and 6.9mm endoscopic.

Complications
One case suffered nerve root outer membrane injury in the 6.9mm endoscopic (1 of 16, 6.25%). However, there was no cerebrospinal uid leakage during operation and no neurological deterioration postoperatively. There were no other severe surgical complications in both groups, such as carotid artery injury, recurrent laryngeal nerve injury, esophageal injury or infection. None of the 28 patients experienced recurrences in the follow-up.

Clinical Outcomes
All the 28 patients completed the follow-up visits. Two of 28 patients, one case in each group, had no signi cant pain relief at 12 months follow-up. The VAS scores ( Figure 13) and the modi ed MacNab criteria ( Figure 14) of 28 patients were evaluated pre-and postoperative. There were no signi cantly different on mean VAS score or outcomes evaluated using the modi ed MacNab criteria between 3.7mm endoscopic and 6.9mm endoscopic. In addition, no difference was found in complication between the two groups (P>0.05).

Follow-up
For all patients, follow-up was performed at 1 day, 1, 3, 6 and 12 months postoperatively. The VAS scores and the modi ed MacNab criteria were performed pre-and postoperatively to evaluate clinical outcomes.
Cervical CT or MRI was performed on all the patients during follow up period.

Discussion
In 1944, Spurling et al. rst described the effectiveness of p-PECD for the treatment of cervical foraminal stenosis which induced by a lateral CDH or osteophytes (32). Studies proved that p-PECD is an effective treatment for cervical diseases and its inner diameter of working cannula ranges from 3.7 mm to 6.9 mm (17,22,28). In our opinion, different diameters of the working cannula may lead to different surgical e ciency. However, no studies were performed to compare the clinical outcomes of 3.7mm endoscopic and 6.9mm endoscopic of p-PECD in CDH patients. In this study, we described the clinical results of 28 consecutive patients diagnosed with unilateral CDH who performed p-PECD using the 3.7mm endoscopic or 6.9mm endoscopic.

Anaesthesia
Studies suggested that local and general anaesthesias are the effective strategies of PECD (17,22,23,33). Wan et al. (17) claimed that local anaesthesia in selected CDH patients is a promising and feasible alternative. However, local anaesthesia still has unavoidable shortcomings, such as uncomfortable and psychentonia during operation. Moreover, if the patient is awake, hearing the voice of the surgical instrument may cause elevated blood pressure, increased heart rate, and poor surgical experience (17). General anaesthesia has been described in several previous studies, which could offer patients a comfortable experience during p-PECD surgery (22,23,33).
In the present cohort, to minimize their intraoperative anxiety and pain and to attain their better cooperation, general anaesthesia was carried out in all patients. Besides, INM technology were used in this study to prevent iatrogenic neurological deterioration intraoperatively. The detailed method refers to Yu et al. (17,34). No nerve compromise was observed in both groups postoperatively and we attribute these positive results to reasonable choice of anesthesia method and application of INM.

Clinical Results
The mean hospital stays of traditional posterior foraminotomy or ACDF in China usually more than seven days (23). In our study, the mean hospitalization of 6.9mm endoscopic and 3.7mm endoscopic were 5.1 (from 2 to 8) and 4.8 (from 3 to 6) days respectively, and both groups had an improvement compared with China average results. Since all the surgeries in this study were performed under general anesthesia, it took about 2 days to complete the preoperative examination and the assessment of the general condition in order to meet the standard of general anesthesia. Usually, patients were discharged 2 days after postoperative observation. Thus, the total length of hospital stay was about 5 days. However, there was no signi cant difference between 3.7mm endoscopic and 6.9mm endoscopic in the average hospital stay period (P>0.05). The longer operative times were required in 3.7mm endoscopic (76.5min) than 6.9mm endoscopic (61.5min), we suggest this result may because of the small-diameter working cannula can only accommodate smaller-diameter endoscopic instruments, such as RF probe, forceps and drill, which obviously limits the e ciency of "V" point identi cation, overlying soft tissue removal and laminoforaminotomy.
On the basis of previous surgical experience (23,35,36), the average VAS score after surgery was signi cantly lower for both endoscopes, however, the difference in the average VAS scores between 3.7mm endoscopic and 6.9mm endoscopic was not signi cant (P>0.05). Meanwhile, considering to the modi ed MacNab criteria, the proportion of having a satis ed result (excellent or good recovery) improved during follow-up visit in both endoscopes, nevertheless, the difference between 3.7mm endoscopic and 6.9mm endoscopic was not signi cant (P>0.05). Therefore, the clinical outcomes of both endoscopes were similarly effective.
Operation Technique ( Table 4) Identi cation of "V" point Identi cation of V-point is an extremely critical operation step, which dominates the success or failure of the p-PECD surgery. Furthermore, accuracy and rapid con rming V-point can provide su cient con dence for the surgeon to proceed with the next step. In our study, identi cation of V-point was easier in the 6.9mm endoscopic than 3.7mm endoscopic (18.608±3.7607min vs. 11.256±2.7161min, p 0.001), which may be attributed to a large diameter of working cannula in the 6.9mm endoscopic.

Potential of spinal cord injury
In this study, neither the 3.7mm nor 6.9mm endoscopic had a surgical complication of spinal cord damage. However, our correspondence author argues that 3.7mm endoscopic has a higher risk of spinal cord injury than 6.9mm endoscopic. The minimal working cannula of 3.7mm endoscopic has a danger of trapping into the spinal canal through the iatrogenic hole and damaging the spinal cord. Meanwhile, 6.9mm endoscopic has a wide enough outer surface of working cannula to prevent negligently inserting into the spinal canal, which increases the safety of the operation. This idea was also agreed by Lin et al. (20), who suggested that increasing outer diameter of the working cannula can reduce the risk of spinal cord injury.

Anterior decompression
3.7mm endoscopic is better than 6.9mm endoscopic for anterior decompression of the intervertebral foramen due to its smaller outer diameter of working cannula which reduces the compression of the spinal cord. In contrast, through the delta working channel, which has a large inner diameter, may lead to spinal cord injury. Possiblity of spinal cord injury High Low * Represents the author's own view of the differences in surgical characteristics between groups 3.7mm and 6.9mm.

Complications
Surgical related complications, including headache, neck pain, dural damage, nerve roots or spinal cord injury, seizures or neurological deterioration due to the highly increased cervical epidural pressure by continuous saline irrigation, intraoperative bleeding or postoperative epidural bleeding, instability caused by surgical and infections, could happen in p-PECD for CDH patients (13,23). In the present cohort, nerve root outer membrane was torn in one case in the 6.9mm endoscopic, but there was no cerebrospinal uid leakage during operation and no neurological deterioration postoperatively. No other surgical complications were observed in both groups. The overall incidence of surgical complication in our study was 3.7% (1/28), and this result is similar to previous studies (22) (23).

Limitation
Despite positive clinical outcomes were achieved in this study, there were still many limitations. The limitations of our study include the small sample sizes, the lack of randomization, the use of single surgeon, the de ciency of multicenter research and the comparably short-term follow-up period. Therefore, multicenter randomized controlled trials with large sample size and long-term follow-up visit should be established.

Conclusions
In the present study, the clinical outcomes between the two endoscopes did not differ signi cantly. 3.7mm endoscopic and 6.9mm endoscopic have their respective advantages. When considering "V" point determination, overlying soft tissue removal and spinal cord injury prevention, 6.9mm endoscopic may be preferable. However, 3.7mm endoscopic may be a better option for anterior decompression of the intervertebral foramen. Overall, p-PECD, including 3.7mm endoscopic and 6.9mm endoscopic, is reliable alternative management of CDH.

Consent for publication
We have obtained consent to publish from the participant to report individual patient data.
Availability of data and materials The datasets used during the current study are available from the corresponding author on reasonable request.

Competing interests
The authors declare that they have no competing interests  Figure 1 The patient was placed in a prone position. The surgeon and assistant manipulate on the same side of the pathology, while the endoscopic display was placed on the opposite side. SBIL indicates superior border of the inferior laminae; IBUL, inferior border of the upper laminae; LF, ligamentum avum.

Figure 2
Intraoperative uoroscopic images during p-PECD with 6.9mm endoscopic. (A) The lamino-facet junction is localized on anteroposterior view; (B) obturator is placed in the "V" point; (C) the working cannula is located over the "V" point.         (A) The mean time of "V" point identi cation in 3.7mm endoscopic and 6.9 mm endoscopic was compared; (B) The mean time of removed overlying tissue in 3.7mm endoscopic and 6.9 mm endoscopic was also compared.

Figure 13
The course of the arm and neck pain in both groups, which was rated using the mean visual analogue scale values.