Patients with bony stenosis of a single-segment cervical nerve root canal; patients who had no remission of pain and numbness after 3 months of conservative treatment; and patients with repeated episodes or worsening or intolerable pain.
Patients with unstable lesions, severe malformations or spinal infections on imaging; patients with cervical spondylotic myelopathy or peripheral neuropathy; and patients with symptoms inconsistent with preoperative imaging.
General characteristics of the patient
In this study, 30 patients were enrolled, including 13 males and 17 females; the age ranged from 43 to 74 years, with an average of 54.3 years. Cervical lesions: C4-C5, 4 cases; C5-C6, 14 cases; C6-C7, 11 cases; C7-T1, 1 case. All patients had symptoms of radiculopathy, numbness, allergies and sensation caused by unilateral nerve root canal stenosis. Among the 30 patients, 26 were positive for the Eaten sign and 27 were positive for the Spurling sign. All patient imaging showed a significant narrowing of the corresponding nerve root canal.
Measurement of stenosis length
All patients underwent CT and MRI scans of the affected segments before surgery. The nerve root diameter “a” of the affected segment was obtained from the preoperative MRI images, and the diameter “a” of the nerve root measured in Fig 1a. 1was set as the reference. On the preoperative CT images, a rectangle with the width equal to the nerve root diameter was drawn to trace the nerve root. One side of the rectangle was set close to one side of the facet joint. The rectangular tracing intersected with the Luschka joints at two points: B and C. The distance between the two points was defined as the length “d” of the narrowing site, as shown in Fig 1b.
Under general anaesthesia with tracheal intubation, the patient was placed in the prone position with their head secured in the neutral position and their neck slightly tilted forward to maximize the cervical intervertebral space. The patient’s arms were secured on the sides of their body and retracted properly towards the feet to avoid affecting the positioning of the cervical vertebrae during intraoperative C-arm radiography. The affected segment was identified and marked via the lateral view of the C-arm radiography. The operative field was routinely disinfected and draped. The Kirschner wire was placed based on the marked point into the junction of the upper and lower lamina medial to the facet joint of the diseased segment. After positioning in the C-arm radiography, a 0.7-cm-long skin incision was made, and the dilator and working trocar were placed over the Kirschner wire (Fig. 2). The Kirschner wire and the dilator were withdrawn. The endoscope system was placed. The electrical hook was used to slowly separate some attachments of the ligamentum flavum medial to the upper and lower lamina and the facet joint in Fig. 3a. Then, a high-speed grinding drill was used to remove the outer cortical bone and cancellous bone of the medial edge of the facet joint and part of the lamina of the upper and lower vertebral body. To avoid injury to the nerve root or spinal cord, the remaining contralateral cortical bone was gradually removed in steps by the hook and rongeur, thus establishing a safe operative field in Fig. 3b. Part of the ligamentum flavum and soft tissue in the interlaminar space were separated carefully and removed. At this time, the compression of the nerve root could be observed under endoscopy, and the tension of the nerve root could also be evaluated by probing with the hook. Next, the medial side of the facet joint as well as the upper and lower laminae could then be further grinded off until there was no significant tension in the nerve roots. Before grinding, the hook was used to explore the nerve root and surrounding tissue to identify the adhesion site in order to avoid injuring the nerve root. After grinding was complete, the hook was used to examine the nerve root again to confirm no compression on the nerve root (Fig. 3c-f). After complete decompression and confirmation of no active bleeding, the endoscope system was withdrawn, and the incision was closed.
Measurement of the length of the decompression zone
After surgery, CT imaging was repeated. The distance between the ends of the bone removal area on the medial edge of the facet joint was measured on CT; this distance is the length of decompression “e” (Fig. 4).
The sphericity test showed that the chi-squared approximation of the VAS score was 30.429 (P ≤ 0.05). The Greenhouse-Geisser correction was used for the analysis, and the results were as follows: F (2.564, 74.367) = 523.963 and P ≤ 0.05, indicating that the differences between the groups were statistically significant. The chi-squared approximation of the NDI score was 257.138 (P ≤ 0.05). The Greenhouse-Geisser correction was used for the analysis, and the results were as follows: F (1.050, 30.447) = 276.364 and P ≤ 0.05, indicating that the differences between the groups were statistically significant.