3D printed guiding device assisted bilateral crossing cervical microendoscopic laminoplasty for cervical spondylotic myelopathy: a study protocol for a randomized controlled trial


 Background

Cervical microendoscopic laminoplasty (CMEL) is an important surgical method for the treatment of cervical spondylotic myelopathy (CSM). It has the advantages of minimal incision and quick recovery, meanwhile, it also has disadvantages, such as its tedious to install the implants and long operative time. This study proposes a new method of 3D printed guiding device assisted bilateral crossing cervical microendoscopic laminoplasty for CSM, which can further improve the accuracy and efficiency of surgery and shorten the operative time.

Methods/design

Patients who met the surgical criteria will be randomly divided into experimental group and control group. The experimental group will be treated with 3D printed guiding device assisted bilateral crossing cervical microendoscopic laminoplasty. The control group was treated with CMEL. The primary outcomes will be incision length, operative time, intraoperative blood loss and adverse events. The secondary outcomes will be the Japanese Orthopedic Association (JOA) score, the visual analogue scales (VAS) for neck and shoulder pain, cervical curvature index (CCI), range of motion (ROM), the sagittal diameter of the spinal canal, and the healing of the grooves. These parameters will be evaluated preoperatively, intraoperatively, 1 week postoperatively, as well as 3, 6, 12, 24 months postoperatively.

Discussion

This study was the first randomized controlled trial to compare 3D printed guiding device assisted bilateral crossing cervical microendoscopic laminoplasty with CMEL. The results of this study may facilitate the development of surgical treatment of cervical spondylotic myelopathy.


Background
Surgical treatment of CSM has been developed for decades [1]. Cervical laminoplasty is an important method for the treatment of CSM. The methods include open-door laminoplasty and bilateral open-door laminoplasty. Although laminoplasty had a satisfactory long-term effect, postoperative neck and shoulder pain, decrease in lordosis and ROM is observed in many patients [2][3][4].
However, there are still some di culties in installing titanium mini-plate under the endoscope and achieving the retroposition of the spinous process-ligament complex: (a) Poor control accuracy: the retroposition of the spinous process-ligament complex relied on intraoperative manual lifting, which was inconvenient to precisely control; (b) Di culty in operation: because the titanium mini-plate must be matched with the surface of lamina, the sliding grooves of the bilateral mini-plate cannot be parallel to the sagittal plane of the human body, which led to some troubles in the retroposition of the spinous process-ligament complex in this way; (c) The operation was cumbersome: because the mini-plate had a certain length and was installed on the same side of the operation, which led to a larger incision. The structure of the spinous process was big on the top and small on the bottom, which led to twist screws on the spinous process side was di cult, and the operation was laborious and time-consuming; (d) Endoscopic operation was prone to skew: due to the anatomical markers cannot be used for positioning under direct vision, the long groove-shaped decompression was prone to skew, which may miscut the bone of the screw installation site, resulting in the di culty in the installation of titanium mini-plate.
These factors had seriously affected the promotion of this minimally invasive technology. Therefore, improving surgical precision, reducing surgical trauma, reducing the di culty of the operation, and improving the e ciency of surgery have become important problems to be solved urgently.
To solve the above operational problems, we have developed a new surgical operating device for the bilateral crossing cervical microendoscopic laminoplasty (the patent application number is CN201910086277). Based on surface features of the lamina, a 3D printed guiding device is used for locating and navigating.
For all we know, there are no randomized controlled trials of 3D printed guiding device assisted bilateral crossing cervical microendoscopic laminoplasty versus CMEL for CSM. In this trial, we will conduct a randomized controlled trial (RCT) to compare 3D printed guiding device assisted bilateral crossing cervical microendoscopic laminoplasty with CMEL to demonstrate this new surgical method is safer, easier and more effective.

Methods/design Study design
The study is a single-centre, prospective, randomized controlled trial, the objective of which is to evaluate the e ciency and safety of 3D printed guiding device assisted bilateral crossing cervical microendoscopic laminoplasty for CSM. A total of 256 patients will be randomly divided into experimental group and control group in a 1:1 ratio. The experimental group will be treated with 3D printed guiding device assisted bilateral crossing cervical microendoscopic laminoplasty. The control group was treated with CMEL. The ow chart of the study process is as follows in Fig. 1.
This trial is reported in accordance with the Standard Protocol Items: Recommendations for Intervention Trials (SPIRIT) guidelines [14] (Fig. 2, Additional le 1)

Ethical approval
The trial meets the requirements of the Helsinki Declaration formulated by the World Medical Association, and it has been reviewed and approved by the institutional ethics review board of the First A liated Hospital of Zhengzhou University (Approved number: KY-2019-LW-049). All participants will be required to sign an informed consent. The protocol had been registered in the Chinese Clinical Trial Registry (ChiCTR). In 2007, ChiCTR was appointed to be the representative registry of China and joined the WHO ICTRP. The protocol number is ChiCTR1900023100.

Participants
The study is a RCT conducted at the First A liated Hospital of Zhengzhou University. Baseline characteristics of patients, such as age, gender, time of onset, symptoms of neck and limbs will be provided. Treatment decisions are made based on baseline characteristics. Surgical treatment is the best choice for patients with ineffective conservative treatment, aggravated pain, and even muscle paralysis. 5. Unwilling to sign informed consent for the study.

Recruitment
The trial recruited the inpatients of the Department of Orthopaedics, the First A liated Hospital of Zhengzhou University, and distributed the research plan lea ets to the patients. After reading the lea ets, the patients and their families voluntarily decided whether to participate in the trial. Screening was conducted according to inclusion and exclusion criteria, patients could only be included in the study after signing the informed consent.

Randomisation and blinding
Patients will be randomly divided into an experimental group and a control group in a 1:1 ratio. It is impossible to blind the participants and surgeons, only the result evaluators are blinding, and these evaluators are not involved in patient management. To prevent bias, the patient's randomization process will be performed by a third party.

Sample size calculation
We carried out a power analysis to evaluate the required sample size to show safety with a power (1-β) of 0.8 and α of 0.05. According to the relevant literature [5], the excellent and good rate of the control group was 84.4%. We hypothesise that the excellent and good rate of the experimental group could be 95%. We carried out a two independent samples t-tests using PASS (Power Analysis and Sample Size), and obtained a result of 128.

Surgical procedures
Experimental group Before the operation, the patient underwent a CT scan of the cervical spine by 64 slice spiral CT (Siemens, Germany). After the scan, the original data was exported and stored in the standard STL format, and the STL format le was imported into E3D Medical Software for 3D reconstruction. The technique of 3D reconstruction which extracts the posterior surface features of the cervical spine, may help perform preoperative simulation on the obtained 3D model and determine some important information, such as the position of the decompression grooves, the entry point and depth of the lateral mass screws, the horizontal distance between the decompression groove and the entry point of the lateral mass screw on the same side, also the entry point and angle of the connecting rods pass through the root of the spinous process. A guiding device includes two guiding tubes, two guiding grooves and two aiming devices. The guiding tubes, the guiding grooves and the aiming devices were designed by using E3D Medical Software (Fig. 3), then printed by 3D printing technology, ensuring that the customized ends of the bilateral guiding tubes were perfectly matched with the bilateral lamina surface. Four indicators can be installed on the outside of the guiding tube. When customized ends of the bilateral guiding tubes were perfectly matched with the surface of lamina, the protruding height of the four indicators was identical.
After initiation of general anesthesia, the patient was placed in a prone position, the head was xed. Under uoroscopic guidance, the targeting level was con rmed. Two lateral incisions approximately 20mm in length were made at the targeting level. Then, the dilator was inserted into the bilateral incision in order, until it contacted the lamina. The dilator was removed after the guiding tube was inserted. The two guiding tubes were connected with the bilateral guiding grooves. The customized ends of the bilateral guiding tubes were perfectly matched with the surface of lamina by adjusting the guiding tubes, and then con rmed by the indicators (Fig. 4A). The guiding tube on the one side was removed, then the endoscope was inserted into the guiding tube on the contralateral side. Endoscopic partial laminotomy was initiated using a high-speed drill to roughen the conjunction of the lamina and the lateral mass, and the ventral cortex of the lamina was completely removed by using a 2mm Kerrison laminectomy rongeur.
Then a groove of approximately 2 to 3mm wide was made at the junction (Fig. 4B). The guiding tube was removed towards the lateral side by a certain distance(this distance is the horizontal distance between the decompression groove and the entry point of the lateral mass screw on the same side). Meanwhile, with the help of the working sleeve, the Magerl technique was used to x the lateral mass screw at the lateral mass (Fig. 4C) [15]. The guiding groove was removed, the aiming device was installed on the guiding tube in the next, then the connecting rod passed through the root of the spinous process by using the aiming device and was xed at the lateral mass screw in the last (Fig. 4D). Thereafter, the working sleeve was removed, the two guiding tubes were reconnected with the bilateral guiding grooves, and the same process was repeated on the contralateral side. Then two working sleeves were connected with the bilateral lateral mass screws. By adjusting the retroposition locking mechanism of the lateral mass screw to make the bilateral connecting rods retroposition 1 to 3mm on the sagittal plane, therefore, the spinous process-ligament complex and the deep extensor muscle group also retroposition 1 to 3mm on the sagittal plane (Fig. 4E,F).
After all of the above steps had been completed, the surgical cavity was ushed, the bleeding was tightly stopped, and a drainage tube was placed. Finally, the fascia layer and skin incision were closed by using standard techniques.

Control group
After initiation of general anesthesia, the patient was placed in a prone position, the head was xed.
Under uoroscopic guidance, the targeting level was con rmed. Then, a midline skin incision approximately 20mm in length was made at the targeting level. Bilateral fascial layers were incised and a hole was drilled in each spinous process. Silk suture was used to thread through the hole, and prepared for suspension of the posterior column structure. Then, the dilator was inserted into the left/right side of the spinous process in order, until it contacted the lamina. The dilator was removed after the working sleeve of endoscope was placed on the surface of the lamina through it. Endoscopic partial laminotomy was initiated using a high-speed drill to roughen the conjunction of the lamina and the lateral mass, and the ventral cortex of the lamina was completely removed by using a 2mm Kerrison laminectomy rongeur.
Thereafter, a groove of approximately 2 to 3mm wide was made at the conjunction. The working sleeve of endoscope could be swung cranially and caudally, and the small incision was long enough for three continuous levels simultaneously. The silk suture was pulled dorsally, then under its protection, the same process was repeated on the contralateral lamina. After decompression of the contralateral lamina was completed, the spinous process-ligament complex and the deep extensor muscle group retroposition 1 to 3mm on the sagittal plane by pulling the silk suture. Thereafter, the medial side of the mini-plate was xed at the root of the spinous process with a small screw, the lateral side of the mini-plate xed at the lateral mass using the Magerl technique.
After all of the above steps had been completed, the surgical cavity was ushed, the bleeding was tightly stopped, and a drainage tube was placed. Finally, the fascia layer and skin incision were closed by using standard techniques. . The healing of the grooves.

Adverse event
Researchers should record the adverse events that occurred after cervical spine surgery and report them to the First A liated Hospital of Zhengzhou University for a time limit of 24 hours. Speci c adverse events include neck and shoulder pain, incision infection, spinal cord injury, nerve root injury, the dural sac tear, major vascular injury, screws shedding and loosening, rods shedding and loosening. If these adverse events occur, the patient should be treated accordingly.

Assessment of outcome
The clinical outcome was measured by the visual analog scale(VAS) score (0=no pain, 10=worst imaginable pain) of axial neck pain (neck, neck or shoulder pains). The preoperative and postoperative neurological condition was evaluated by the Japanese Orthopaedic Association (JOA) score [16]. The excellent rate proposed by the method by Hirabayashi [17] was calculated using the following formula: Recovery rate (%)=[(postoperative JOA score preoperative JOA score) /(17 preoperative JOA score)]×100. The criteria for e cacy grading are: excellent, recovery rate ≥ 75%; good, 50% to 74%; fair, 25% to 49%; poor, <25%. The incidence of adverse reactions was calculated and used to evaluate the safety of the procedure. The incidence of adverse reactions = the number of patients with adverse reactions / the total number of patients × 100%.

Radiographic evaluation
Outpatient follow-up was performed regularly at 3, 6, 12, 24, and 36 months after surgery. All patients underwent cervical spine radiographs at each follow-up (including lateral, neutral, exion, and extension views). Cervical curvature index (CCI) and cervical range motion (ROM) were calculated. The CCI was calculated by measuring the angle between the upper edge of C3 and the lower edge of C7 on the lateral radiograph of the cervical spine. The ROM was calculated by measuring the difference of included angle between C2 and C7 vertebral lower margin on lateral X-ray lms of cervical hyperextension view and hyper exion view. CT scan was used to evaluate the degree of spinal canal enlargement and bony unions of the grooves. MRI was used to evaluate the change of signal intensity and the decompression condition of the spinal cord.

Data management
Data were collected in the form of a Case Report Form (CRF) and summarized in a uni ed form. Speci c data include the following aspects: demographic information, disease diagnosis, disease concomitant conditions and adverse reactions. These data are recorded and saved as electronic les by using Epidata data processing software. Research data of this clinical trial are kept by the First A liated Hospital of Zhengzhou University. Only researchers have the right to inquiry the documents of the database. The Independent Data Monitoring Committee (IDMC) supervises and manages data throughout the clinical research process. Professional statisticians will make a statistical analysis of the database les and get the statistical analysis results. The main researchers of the trial then write the research report based on the statistical results.

Statistical analysis
All statistical analyses are performed with SPSS V.21.0 statistical software (SPSS, Inc, Chicago, IL). All data are expressed as the mean ± standard deviation. The difference of incision length, operative time, and intraoperative blood loss between the two groups will be analyzed using a two independent samples t-tests (α= 0.05). JOA score, VAS score, CCI, ROM and the sagittal diameter of the spinal canal of preoperatively, 1 week postoperatively, as well as 3, 6, 12, 24 months postoperatively will be analyzed using a repeated-measures analysis of variance. A value of P < 0.05 is considered statistically signi cant. This procedure can effectively enlarge the spinal canal while retaining most of the normal bone-ligament anatomical structure of the cervical spine. After full decompression, two lateral mass screws were vertically twisted into the lateral mass on both sides. The connecting rod was twisted into the root of the spinous process through the aiming device and then xed with the lateral mass screw. By adjusting the retroposition locking mechanism of the lateral mass screw to make the bilateral connecting rods retroposition 1 to 3mm on the sagittal plane, thereby the spinal canal was enlarged. With the help of the bilateral connecting rods, the spinous process-ligament complex and the lateral mass screws formed a stable triangular structure in posterior column of this level which further reducing the risk of postoperative spinal instability.

Discussion
Indications of 3D printed guiding device assisted bilateral crossing cervical microendoscopic laminoplasty are as follows: (a) Congenital spinal canal stenosis; (b) Spinal canal stenosis caused by ossi cation of the posterior longitudinal ligament, hypertrophy of ligamentum avum, and proliferation of articular process; (c) The dura mater is compressed by multi-level cervical disc herniation; (d) The anterior cervical operation cannot alleviate symptoms. Contraindications include patients with cervical kyphosis, cervical instability, cervical spondylolisthesis over 1 grade and severe osteoporosis.
This trial was the rst randomized controlled trial to compare 3D printed guiding device assisted bilateral crossing cervical microendoscopic laminoplasty with CMEL. The purpose of the trial was to introduce a new minimally invasive laminoplasty technique and to demonstrate that this new surgical method is safer, easier and more effective.    The designed guiding tubes, guiding grooves and aiming device in E3D Medical Software.

Figure 4
Diagrammatic sketch of the 3D printed guiding device assisted bilateral crossing cervical microendoscopic laminoplasty. (A) The two guiding tubes were connected with the bilateral guiding grooves. After the two guiding tubes passed through the incisions on both sides, customized ends of the guiding tubes were perfectly matched with the surface of lamina by adjusting the guiding tube, and con rmed by the indicators. (B) The guiding tube on the one side was removed. Then the endoscope was inserted into the guiding tube on the contralateral side. Endoscopic partial laminotomy was initiated using a high-speed drill to roughen the conjunction of the lamina and the lateral mass, and the ventral cortex of the lamina was completely removed by using a 2mm Kerrison laminectomy rongeur. (C) The guiding tube was removed towards the lateral side by a certain distance, then the lateral mass screw was xed at the lateral mass with the help of the working sleeve. (D) The aiming device was installed on the guiding tube, then the connecting rod was passed through the root of the spinous process by using the aiming device, and it was xed at the lateral mass screw. (E) The same process was repeated on the contralateral side. Then two working sleeves were connected with the bilateral lateral mass screws. By adjusting the retroposition locking mechanism of the lateral mass screw to make the bilateral connecting