Introduction

Cervical spondylotic radiculopathy (CSR) is closely associated with the posterior lateral wall of the uncinate process. Uncinate process hyperplasia of osteophytes leads to foraminal stenosis and compression of the cervical nerve front, resulting in clinical manifestations of neck, shoulder, and upper-limb pain1. Some scholars believe that complete or partial removal of the uncinate process during anterior cervical discectomy and fusion (ACDF) can achieve complete decompression of the intervertebral foramina with better postoperative efficacy2. However, during conventional ACDF, after complete removal of the intervertebral disc, surgeons often decompress the cervical nerve by excising the uncinate process with the naked eye, which carries the risk of cervical nerve injury. With the continuous progress in spinal endoscopy technology, endoscopic technology has improved surgical safety and efficiency during resection of the uncinate process to a certain extent3,4. However, there is still no consensus on the scope of surgery when combined with uncinate process resection, and surgeons often rely on experience to remove the uncinate process endoscopically. Excessive removal of the uncinate process increases irreversible cervical nerve damage and postoperative cervical instability, and too little removal of the uncinate process results in incomplete decompression5,6,7,8,9,10. Precise removal of the uncinate process to achieve complete decompression of the cervical nerve has become an urgent problem.

Aleem, I. S. et al.11 reported that the posterior endplate valley (PEV) is the midpoint of the posterior superior corner of the vertebral body, which is the lowest point of the valley (or “smile”) in the coronal plane. The PEV can be used as a reliable intraoperative guide for ACDF. Moreover, the lateral margin of the vertebral surface surrounds the epiphyseal ring12. On the basis of this view, it is reasonable to consider the definition of the PEV as the lowest point of the epiphyseal ring posterior margin (LPER) under endoscopy and regard the LPER as a significant symbol for safe decompression of the cervical nerve. However, no studies have described the safety indices of quantitative resection of the uncinate process through the LPER using spinal endoscopy. Mimics software combined with CT myelography (CTM) was used to observe the positional relationship between the LPER, the cervical nerve and the isthmus of the vertebral pedicle in 3D images more intuitively and accurately and to determine the safe distance for removing the uncinate process. Our study will not only help achieve complete decompression of the cervical nerve but also make the operation more accurate and safer, thereby providing a theoretical basis for clinical practice.

Participants and methods

Participants

A total of 26 patients with CSR who were diagnosed and treated at the Affiliated Hospital of Binzhou Medical College from December 2016 to April 2019, including 12 males and 14 females, were selected as the study subjects. The ages ranged from 32 to 72 years, with a mean of 50.58 ± 7.40 years. The inclusion criteria for patients were as follows: (1) were diagnosed with CSR, characterised by neck and shoulder pain accompanied by upper-limb anaesthesia, upper-limb paraesthesia and other nerve compression symptoms; (2) imaging findings showing compression of the cervical nerve by the foramen intervertebral anteromedial bony structures of the foramina or/and intervertebral discs that were consistent with symptoms and signs; and (3) consent to cervical myelography. The exclusion criteria were as follows: (1) had congenital or acquired cervical deformity; (2) had a history of cervical spine surgery; (3) had cervical spine tumours; and (4) had cervical neurodevelopmental abnormalities, cervical fractures, or other anatomical changes. This study was approved by the Medical Ethics Committee of the Affiliated Hospital of Binzhou Medical College, [2023] No. Lunshenzi (KT-137).

Imaging methods

Patients were placed in a lateral position for subarachnoid puncture; approximately 10 mL of cerebrospinal fluid was allowed to flow naturally, and the same amount of iohexol (GE Healthcare Shanghai Co., Ltd.) was injected. After lying flat for 1.5 h, a cervical spine CT scan was performed on a GE 64-row spiral C.T. (manufactured in the United States). During the scan, the patients lay flat, the spine remained neutral, the back of the neck was raised, the same position that would be used during surgery was assumed, and the thin layer scan width was 0.625 mm. The scanned data were stored in DICOM format. The data were imported into Mimics 21.0 software (manufactured in Materialise, Belgium) to reconstruct the three-dimensional model of the cervical vertebra.

Determination of bone markers

The epiphyseal ring posterior margin is not easily proliferative, showing a middle depression; both sides are raised, and the epiphyseal ring extends laterally to form the uncinate process tip and its medial surface. The lowest point is clearly visible under endoscopy, relatively fixed and smile shaped; thus, it can be used as a bone marker to locate the location of the cervical nerve for safe decompression.

The isthmus of the vertebral pedicle is narrow, and both ends are enlarged, similar to a dumbbell. The isthmus is the narrowest site of the intervertebral foramen and the main site of cervical nerve compression. Thus, the lateral border of the isthmus of the pedicle serves as the limit for complete decompression of the cervical nerve in the pedicle (i.e., the foramen).

LPER and measurement indices were determined in 3D images (Fig. 1a and c). The indices were as follows: the anteroposterior distance from the LPER to the anterior margin of the cervical nerve at the position of the uncinate process tip (ANU), the vertical distance from the LPER to the inferior margin of the cervical nerve at the position of the uncinate process tip (VNU), the vertical distance from the LPER to the cervical nerve inferior margin origin (VNI), the vertical distance from the LPER to the cervical nerve superior margin origin (VNS), the left-right distance from the LPER to the dura mater lateral border (DLB), and the anteroposterior distance from the LPER to the vertebral pedicle isthmus lateral border (PLB).

Fig. 1
figure 1

Schematic diagram of 3D images. (a-c) LPER: the lowest point of epiphyseal ring posterior margin. VNU: the vertical distance from the LPER to the inferior margin of the cervical nerve at the position of the uncinate process tip. VNI: the vertical distance from the LPER to the cervical nerve inferior margin origin. VNS: the vertical distance from the LPER to the cervical nerve superior margin origin. DLB: the left-right distance from the LPER to the dura mater lateral border. ANU: the anteroposterior distance from the LPER to the anterior margin of the cervical nerve at the position of the uncinate process tip. PLB: the anteroposterior distance from the LPER to the vertebral pedicle isthmus lateral border.

Full size image

Measurement indices

On the basis of CT, correlative planes were established at the LPER, cervical nerve origin, cervical nerve ventral side and isthmus of the vertebral pedicle. At the sagittal position, ANU, VNU, VNI and VNS were measured (Fig. 2a and b). At the transverse position, DLB and PLB were assayed (Fig. 2c and d). The side and sex differences in the measurement indices were compared in each segment. The differences in all indices across the different segments were compared. The differences between VNU and VNS were compared in each segment.

Fig. 2
figure 2

Schematic diagram of the CT images. (a, b) At the sagittal position, the white line represents the established plane during the measurement process. ANU: the anteroposterior distance from the LPER to the anterior margin of the cervical nerve at the position of the uncinate process tip. VNU: the vertical distance from the LPER to the inferior margin of the cervical nerve at the position of the uncinate process tip. VNI: the vertical distance from the LPER to the cervical nerve inferior margin origin. VNS: the vertical distance from the LPER to the cervical nerve superior margin origin. (c, d) At the transverse position, the white line is the established plane during the measurement process. DLB: the left-right distance from the LPER to the dura mater lateral border. PLB: the anteroposterior distance from the LPER to the vertebral pedicle isthmus lateral border.

Full size image

Statistical analysis

All the measured indices conformed to a normal distribution, and the measurement indices of sex and side in the same segment were tested via independent sample t tests. One-way analysis of variance was used for comparisons between different segments. Calculated the effect sizes for each measurement index and 95% confidence intervals (CI). The results are expressed as the means ± standard deviations. P < 0.05 was considered to indicate statistical significance. Origin 2021 software was used for mapping. Statistical calculations were performed using IBM SPSS (version 26.0, IBM Corp.).

Results

Measurement outcomes of the ANU, VNU, DLB, VNS, VNI, and PLB in the same segment and the index differences between the two sides

There were no statistically significant differences in the ANU, VNU, DLB, VNS, VNI or PLB of the same segment between the two sides (P > 0.05) (Table 1).

Table 1 Data of the measurement indices of the side in a 3D image.
Full size table

Sex differences in comparative measurement indices

There were no significant differences between the two sides of the same segment (P > 0.05). Therefore, the left and right values of 12 males and 14 females were synthesized into 24 and 28 samples in the same segment for statistical analysis, respectively. There were no statistically significant differences in each index of the same segment according to sex (P > 0.05) (Table 2).

Table 2 Data of the measurement indices of the different sexes in 3D images. The index values are presented as the means ± standard deviations. ANU: the anteroposterior distance from the LPER to the anterior margin of the cervical nerve at the position of the uncinate process tip. VNU: the vertical distance from the LPER to the inferior margin of the cervical nerve at the position of the uncinate process tip. DLB: the left-right distance from the LPER to the dura mater lateral border. VNS: the vertical distance from the LPER to the cervical nerve superior margin origin. VNI: the vertical distance from the LPER to the cervical nerve inferior margin origin. PLB: the anteroposterior distance from the LPER to the vertebral pedicle isthmus lateral border.
Full size table

Comparisons of the critical indices among different segments

There were no significant differences in the side or sex of the 26 patients (P > 0.05). Therefore, every measured index on two sides of 26 patients was synthesized into 52 samples for statistical analysis of ANU, VNU, DLB, VNS, VNI, and PLB. From C3/4 to C7/T1, the differences in the ANU between each segment were not statistically significant (P > 0.05). The VNU gradually increased, whereas the DLB gradually decreased. The PLB first increased but then decreased. The decompression distance at the outlet of the foramen decreased at C6/7 and C7/T1. VNU and VNI, and the differences between each segment were not statistically significant (P > 0.05), indicating that decompression of the lower margin of the cervical nerve at the position of the uncinate process tip could be used to achieve decompression of the lower margin of the cervical nerve origin (Table 3).

Table 3 Comparisons of the CTM measurement indices among the five spinal segments. The values are presented as the means ± standard deviations. ANU: the anteroposterior distance from the LPER to the anterior margin of the cervical nerve at the position of the uncinate process tip. VNU: the vertical distance from the LPER to the inferior margin of the cervical nerve at the position of the uncinate process tip. DLB: the left-right distance from the LPER to the dura mater lateral border. VNS: the vertical distance from the LPER to the cervical nerve superior margin origin. VNI: the vertical distance from the LPER to the cervical nerve inferior margin origin. PLB: the anteroposterior distance from the LPER to the vertebral pedicle isthmus lateral border.
Full size table

Discussion

During the surgical procedure, after the complete removal of the intervertebral disc, the endoscope can clearly visualize the anatomical structures of the posterior margin of the epiphyseal ring and the uncinate process. Under the endoscope, the operations are more precise. This study measured the ANU, DLB, PLB, VNS, VNU, and VNI indices to achieve precise endoscopic resection of the osteophytes on the uncinate process and the upper vertebral body, which reduces the risk of cervical nerve injury compared with conventional ACDF surgery. This approach avoids the blind resection of bone in conventional ACDF surgery and achieves complete decompression of the cervical nerve from the cervical nerve origin to the intervertebral foramen outlet. Statistical analysis revealed that there was no significant difference in the measurement indices between sex and side, and the surgeon did not need to take these factors into account during nerve exploration. From C3/4 to C7/T1, the ANU was (1.34 ± 0.79–1.52 ± 0.80) mm, and there were no differences among the five segments. There were significant differences in DLB (8.19 ± 0.80–9.69 ± 0.84) mm, PLB (2.35 ± 1.03–0.90 ± 0.46) mm, VNU (2.93 ± 1.32–4.72 ± 2.35) mm, VNS (8.77 ± 2.25–11.35 ± 2.40) mm and VNI (3.13 ± 1.35–5.18 ± 2.52) mm among the five segments (P < 0.05). In the same segment, there was no significant difference between VNU (2.93 ± 1.32–4.72 ± 2.35) mm and VNI (3.13 ± 1.35–5.18 ± 2.52) mm (P > 0.05). To our knowledge, these findings are the first reports of the use of endoscopic LPER as a bone marker to locate the cervical nerve, define the decompression range in the foramen, and precisely remove the uncinate process.

Some scholars have divided the cervical neural tube into the medial region (pedicle), intermediate region (vertebral artery), and lateral region (transverse ridge)13. The medial region (i.e., the foramen) is thought to play an essential role in the aetiology of CSR. In the anterior part of the foramen, the Luschka joint is involved mainly in cervical nerve compression14,15,16. The cervical nerve runs from posterosuperior to anterolateral to the uncinate process. Therefore, the precise excision of the osteophytes on the uncinate process and the superior vertebral body from the outlet of the intervertebral foramen to the cervical nerve origin under endoscopy is key to adequate decompression of the cervical nerve.

During the surgical procedure, the standard steps include sequentially incising the skin, subcutaneous tissue, and other soft tissues. After the surgical segment is identified, the nucleus pulposus is removed using a rongeur, and an endoscopic system is installed to observe and expose the superior and inferior endplates. Under endoscopy, the LPER serves as a bone marker for decompressing the cervical nerve. The first key to success lies in decompression at the intervertebral foramen outlet. From the LPER, decompression of the ventral side of the cervical nerve at the intervertebral foramen outlet is achieved by moving forwards the horizontal distance corresponding to the ANU with a curette. The procedure then continues from the LPER, moving forwards the horizontal distance corresponding to the PLB to reach the lateral margin of the pedicle isthmus. Before decompression, the drill diameter can be used as a reference marker. After drilling a hole in the medial wall of the uncinate process, a curette is used to slowly remove the osteophytes of the uncinate process, and the osteophytes are scraped outwards to the lateral wall of the uncinate process, thoroughly completing the decompression at the intervertebral foramen outlet. The second key to success is decompression at the cervical nerve origin. After decompression at the intervertebral foramen outlet is complete, decompression continues medially and upwards along the path of the cervical nerve. From the LPER, moving outwards by the horizontal distance corresponding to the DLB, a drill with an appropriate diameter for positioning is selected, and hypertrophic or ossified posterior longitudinal ligaments are removed under the endoscope to decompress the ventral side of the dura mater. Subsequently, from the LPER, moving upwards by the vertical distance corresponding to the VNU, the posterior wall of the uncinate process is ground down to decompress the lower margin of the cervical nerve near the uncinate process tip. Continuing from the LPER, moving upwards by the vertical distance corresponding to the VNI, the osteophytes of the superior vertebral body are removed to decompress the lower margin of the cervical nerve origin. Along the cervical nerve path, moving vertically upwards from the LPER by the vertical distance corresponding to the VNS, decompression is completed at the upper margin of the cervical nerve origin, achieving full decompression of the cervical nerve origin.

First, for ventral cervical nerve decompression at the intervertebral foramen outlet, it is essential to know the ANU. Sung-Ho Kim et al.17. reported that the shortest distance from the posterior end of the uncinate process tip to the nerve root was (1.3–2.0) mm, and the surgeon considered these data as a reference when the anterior margin of the cervical nerve at the position of the uncinate process tip at the LPER was located. This study revealed no difference in ANU from C3/4 to C7/T1. However, we observed that in each segment, the coronal plane of the LPER was either anterior or posterior to the anterior margin of the cervical nerve at the position of the uncinate process tip. Therefore, ventral cervical nerve decompression at the intervertebral foramen outlet can be accomplished by probing the cervical nerve forwards or backwards (1.34 ± 0.79–1.52 ± 0.80) mm from the LPER and excising the posterior uncinate process and the osteophyte behind the vertebral body.

Second, after the process of ventral decompression of the cervical nerve in the foramen, it is still necessary to decompress it to the intervertebral foramen outlet. G. Siouta et al.13 reported that the cervical nerve travels through three areas in the foramen and that the narrowest part of the middle area (that is, the isthmus of the pedicle) is the main compression site of the cervical nerve. The lateral margin of the isthmus of the pedicle can be used as the limit of complete decompression of the cervical nerve in the pedicle (that is, the foramen), and there is no need to continue decompression outwards. Therefore, it is also important to understand the PLB. At C3/4–C5/6, from the LPER constant forwards (2.17 ± 0.98–2.35 ± 1.03) mm, the diameter of the drill can be indicated as a reference mark to remove part of the uncinate process and its osteophytes to the lateral margin of the isthmus of the pedicle. At C6/7–C7/T1, the LPER was kept constant at (0.90 ± 0.46–1.14 ± 0.86) mm forwards or backwards to probe the cervical nerve. The standard drilling specifications of C3/4–C5/6 should be avoided. Excessive bone removal may result in cervical instability. When operating at C6/7–C7/T1, smaller diameter drills or tools such as curettes should be used to slowly remove the osteophytes from the posterior wall of the uncinate process. At the same time, decompression at the intervertebral foramen outlet can be thoroughly completed by slowly grinding through the uncinate process outwardly with a grinding drill. Güvençer, M. et al.18. reported that fibro-ligamentous tissue was observed on the lateral side of the uncinate process surrounding the vertebral artery and the cervical nerve; thus, the fibro-ligamentous tissue was used as a safety barrier during outwards grinding to avoid injury to important structures. Raveendranath V19 reported that the anterior and posterior lengths of the uncinate process were approximately 10 mm. In combination with this observation, only approximately one-fifth of the posterior uncinate process and its osteophytes need to be removed.

Although decompression of the cervical nerve at the pedicle (intervertebral foramen) and the outlet was completed, further decompression was required inwardly and upwards along the path of the cervical nerve to its origin. Therefore, it is critical to know the VNS, VNI, DLB, and VNU.

The intervertebral foramen at the exit was filled, and the cervical nerve was compressed, first inwards for dura mater ventral decompression. At C3/4-C5/6, an appropriately sized drill can be selected to locate the lateral margin of the dura mater by extending (9.44 ± 0.80–9.69 ± 0.84) mm laterally from the LPER, enabling ventral decompression of the dura mater. Studies have shown that as the spinal segment descends, the DLB gradually decreases. For the C6/7 and C7/T1, smaller diameter drills can be chosen to extend (9.05 ± 0.81) mm and (8.19 ± 0.80) mm laterally from the LPER, respectively, to remove the hypertrophic or ossified posterior longitudinal ligaments and achieve decompression under the endoscope. During the decompression process, a cluster of tough ligaments can be observed on the anterolateral side of the cervical nerve lateral to the dura mater. The cervical nerve should be carefully cut and exposed, and upwards internal decompression along its path should be continued.

The study revealed that at C4/5-C6/7, selecting an appropriately sized drill based on the VNU and removing the posterior wall of the uncinate process upward from the LPER by (3.27 ± 1.36–3.66 ± 1.58) mm can effectively decompress the cervical nerve near the uncinate process tip. At C3/4, using a smaller drill to remove osteophytes upward from the LPER by (2.93 ± 1.32) mm helps avoid excessive resection, which could lead to cervical nerve injury and cervical spine instability. Conversely, at C7/T1, a larger drill is required to remove osteophytes from the LPER by (4.72 ± 2.35) mm to ensure thorough decompression of the cervical nerve. Furthermore, this study revealed no difference between VNU and VNI in each segment, indicating that after decompressing the lower margin of the cervical nerve at the uncinate process tip, a slight upward exploration along the cervical nerve pathway can effectively decompress the nerve to the lower margin of the cervical nerve origin.

Finally, after decompression at the lower margin of the cervical nerve origin, a drill with an appropriate diameter based on the VNS. The upper margin of the cervical nerve origin was accurately located. At C3/4, the upper margin of the cervical nerve origin was located from the LPER upwards (8.77 ± 2.25 mm), and the osteophytes on the posterolateral wall of the upper vertebral body were removed. As the segment size decreases, switching to drills with larger diameters is recommended to improve positioning and decompression effectiveness, thereby avoiding incomplete cervical nerve decompression caused by improper instrument selection. At C4/5-C5/6 and C6/7-C7/T1, osteophytes are ground from the LPER upwards to distances of (10.06 ± 2.29–10.58 ± 2.31) mm and (11.32 ± 2.91–11.35 ± 2.40) mm, respectively, ensuring complete decompression of the cervical nerve root origin and achieving the desired surgical outcomes.

There are several limitations of this study. First, the routine use of CTM carries potential risks such as pain at the puncture site, bleeding, cauda equina injury, and cerebrospinal fluid leakage, as it is an invasive procedure. Relevant measures should be taken to prevent these risks. Second, this study was purely theoretical, and the sample size was small. Although CSR patients were included in the study, not all segments were in a pathological state. Essentially, we focused on segments that did not exhibit pathological changes, which may limit the generalizability of the findings. In future studies, we will increase the sample size of segments with pathological conditions. The obtained imaging anatomical data provide a reference for future cadaveric anatomical studies and offer valuable insights for clinical surgery. Next, the study participants were all adults, the vast majority of whom were middle-aged or elderly individuals. In this age group, we assume that age differences have a relatively small impact on the results. However, as there is no direct research supporting this assumption, there is currently a lack of evidence to verify its validity. Finally, regarding the impact of body mass index (BMI), this study did not involve biomechanical factors but focused on exploring the relationship between skeletal structure and neural tissue through imaging. Given the relatively low load on the cervical spine, we considered the influence of BMI on the measurement results to be negligible, but this assumption also lacks empirical data support. With respect to the influence of surgeons’ expertise, all measurements in this study were performed by the same surgeon, with two repeated measurements conducted. Therefore, further exploration of the potential impact of different age groups, BMI ranges, and surgeries performed by different surgeons on patient outcomes will be carried out in the future.

Conclusion

In this study, the positional relationships among the LPER, cervical nerve and isthmus of the pedicle were determined via three-dimensional reconstruction and CTM. With LPER as the bone marker, full decompression of the entire cervical nerve from the origin of the cervical nerve to the outlet of the pedicle was completed forwards, inwards, outwards and upwards, providing a reference for spinal endoscopy-assisted anterior cervical surgery and reducing surgical risk.