INTRODUCTION
Cervical spondylotic radiculopathy (CSR), a prevalent degenerative spinal disorder, manifests as neck and arm discomfort and is typically indicative of nerve root compression secondary to foraminal stenosis [1,2,9,33]. As aging populations become more common, the rate of incidence of this condition has increased [7,30]. Historically, anterior cervical discectomy and fusion (ACDF) has been the cornerstone surgical intervention for the management of cervical disc herniation [23,24,28]. Despite its widespread adoption, ACDF has several drawbacks, including the risk of adjacent segment degeneration, various complications, and significant hospitalization expenses [8,12,29]. Consequently, surgeons have continuously searched for more efficient and safer alternatives to ACDF.
Posterior cervical foraminotomy (PCF) was first reported in 1944 and was subsequently validated for therapeutic efficacy in various studies [3,10,22,26]. PCF is distinguished by its minimally invasive nature, familiarity with surgical anatomy via the posterior cervical route, and preservation of segmental motion. The evolution of spinal endoscopic techniques since the early 2000s has led to the integration of PCF with full-and biportal endoscopy, enhancing the benefits of the procedure with the minimally invasive advantages of endoscopy [11,13,15,17,18,20,21,32]. In 2011, a modified variant of PCF, posterior cervical inclinatory foraminotomy (PCIF), was introduced, employing an inclined approach angle to preserve the facet joint and capsule, diverging from traditional PCF methods [4,14]. This technique was further adapted for biportal and full endoscopies by Song and Lee [25] and Kim et al. [16], respectively, combining the merits of PCF, PCIF, and endoscopy over ACDF, thereby offering a comprehensive surgical solution.
This study builds upon preliminary findings from prior research by Song and Lee [25], the first author, who introduced the biportal endoscopic (BE) technique but was limited to a small sample size of seven cases. By incorporating a larger cohort, this study provides stronger evidence of the technique’s efficacy and safety. Additionally, it presents a novel analysis of the relationship between the inclinatory angle and foraminal widening, offering new insights into surgical optimization. This is the first report to evaluate a significant number of cases of cervical foraminotomy using biportal endoscopy with an inclinatory approach. Through this, the study aims to establish the clinical and radiological benefits of BE-PCIF for CSR.
MATERIALS AND METHODS
Study population and design
Following the approval of the research design by the Institutional Review Board (CHOSUN 2024-03-005), a single-center retrospective study was conducted. Patients diagnosed with CSR who underwent BE-PCIF between April 2020 and April 2023 were included, and relevant data were retrospectively collected from clinical charts and radiologic images. The eligibility criteria were as follows : 1) individuals aged 18 and above diagnosed with one or two levels of unilateral symptomatic CSR; 2) demonstration of clear signs of cervical foraminal stenosis on cervical magnetic resonance imaging (MRI), accompanied by upper extremity radicular pain; and 3) showing no improvement after a minimum of 3 months of conservative treatment. Exclusion criteria comprised the following : 1) discrepancies between MRI findings and clinical symptoms; 2) radiculopathy arising more prominently from root compression due to cervical disc herniation than from degenerative spondylotic foraminal stenosis; 3) accompanying segmental instability; 4) central canal stenosis-induced myelopathy; and 5) other concurrent pathological conditions such as infection, trauma, or tumors.
Surgical procedures
Surgical intervention was performed with the patient in prone position under general anesthesia. The head was stabilized, and the shoulders were retracted using plasters, with the posterior neck area prepped with an antiseptic solution and adequately draped.
Positioned contralaterally to the lesion, the surgeon created two portals under C-arm fluoroscopy guidance; each incision measured 0.5 cm and was aligned vertically along the lateral margins of the spinous process. The cranial portal was established at the level of the upper cervical spinous process corresponding to the location of the lesion, and the caudal portal was established at the level of the lower cervical spinous process, with an interportal distance of approximately 2 cm.
The neck musculature was dissected using serial dilators to create an operative field. A 0° endoscope was introduced via the viewing portal following cannula insertion. A saline irrigation system was set up 2 m above the floor of the operating room to facilitate natural infusion. Triangulation was achieved with the endoscope and surgical instruments at the margins of the superior and inferior laminae and the medial facet joint point (Vpoint), where minor bleeding was managed with a radiofrequency (RF) probe. Following the V-point identification, the surrounding bony structures were drilled away using endoscopic drills to preserve the ligamentum flavum and protect the neural elements during laminotomy.
Continuous drilling around the V-point revealed the caudocranial margin of the ligamentum flavum, allowing the surgeon to discern the contour of the nerve root through a thin layer at the lateral margin of the ligamentum flavum. The decompression boundary was expanded laterally at the foramen using a flexible 3.5-mm diamond burr. Circumferential drilling along the nerve root pathway was supplemented with additional decompression using 1-mm Kerrison punches and small curettes. In cases requiring further decompression, the cranial tip of the superior articular process (SAP) was excised using 2-mm punches or a small chisel.
Upon completion of bony decompression, the ligament flavum was excised directionally toward the nerve root from the thecal sac with vigorous hemostasis during removal to maintain a clear surgical field. Hemostasis around the root origin was promptly achieved by using a small RF ablator to manage venous plexus-related bleeding and ensure an unobstructed operative view.
The final verification of complete decompression was assessed by observing dura mater pulsation, followed by thorough hemostasis. Closure involves subcutaneous suturing and the application of skin tape. Postoperatively, the patients were recommended to use a neck collar for 1 week. The surgical procedure is described in detail in a previous study by the first author of this paper [25]. Supplementary Video 1 presents the detailed surgical scene of an illustrative case.
Data collection
Clinical indicators
Comprehensive clinical data, including age, sex, affected side, surgical level, number of surgical levels, operative time, and follow-up period, were collected for all participants. Neck and upper extremity radicular pain was assessed using the Visual analog scale (VAS) at intervals of 3 days, 1 month, 6 months, and 12 months postoperatively. Patient satisfaction was gauged at the last follow-up using the MacNab criteria.
Radiologic assessment
Radiological measurements were derived from plain radiographs and computed tomography (CT) scans pre-operatively or at specified postoperative periods (6 months for radiographs and 1 month for CT scans), encompassing the following : (1) C2-7 lordosis : the angle between the line parallel to the upper border of the C2 vertebral body and the line parallel to the lower border of the C7 vertebral body on the lateral radiograph (Fig. 1); (2) focal segmental angle : the angle between the line parallel to the upper border of the vertebral body proximal to the involved segment and the line parallel to the lower border of the underlying vertebral body on the lateral radiograph (Fig. 1); (3) disc height : the maximal distance between the lower border of the vertebral body proximal to the involved segment and the upper border of the underlying vertebral body on lateral radiography (Fig. 1); (4) superior articular process distance (SAPD) : the distance between the medial margin of SAP and vertebral body on the axial plane of CT (Fig. 2); (5) isthmic distance (ID) : the narrowest width of the foraminal canal on the axial plane of CT (Fig. 2); (6) facet length : the length of the facet joint on the axial plane of the CT (Fig. 2); and (7) inclinatory angle : the angle between a line parallel to the median line and the line extending from the resected facet joint surface on the axial plane of CT at one-month post-operation (Fig. 2).
The calculated indices were as follows : (1) facet resection rate : value obtained by dividing the reduced facet length by the preoperative facet length; (2) SAPD expansion ratio : value obtained by dividing postoperative SAPD by pre-operative SAPD; and (3) ID expansion ratio : value obtained by dividing postoperative ID by pre-operative ID.
Correlation analysis
In this study, a correlation analysis was conducted to determine whether certain factors influence the inclinatory angle formed by the trajectory of surgical approaches, or conversely, the inclinatory angle influences these factors. Collected demographic factors and radiological indices were the focus of this study.
Statistical analysis
Data analysis was performed using R Statistical Software (version 4.2.2; R Foundation for Statistical Computing, Vienna, Austria). Descriptive statistics were applied to all variables, and correlations were investigated using paired t-tests, analysis of variance, or correlation analysis using Pearson’s correlation coefficient. The significance threshold was set at p<0.05.
RESULTS
Patients’ demographics
This study assessed the outcomes of 101 patients with 162 surgical levels, mainly involving the C5-6 and C6-7 levels. The cohort comprised 46 men (45.5%) and 55 women (54.5%), with an average age of 56.99±9.14 years. The number of patients who underwent two-level surgery (61 patients, 60.4%) was greater than the number of patients who underwent one-level surgery (40 patients, 39.6%) and were slightly more common on the 92 segments of the right side (56.8%) than the 70 segments on the left side (43.2%). The operative duration varied depending on the number of levels addressed, with single-level surgeries averaging 59.0±17.8 minutes, and two-level surgeries 92.3±21.4 minutes. The mean follow-up period was 17.3±4.1 months (Table 1).
Clinical outcomes
Pre-operatively, the mean neck pain score was 7.4±1.2, which significantly decreased to 1.7±1.4 at 3 days postoperatively (p<0.001). This trend of improvement continued at 1 month (0.9±1.2, p<0.001), 6 months (0.5±1.0, p<0.001), and 12 months (0.4±0.9, p<0.001) follow-up. Similarly, the pre-operative mean radiating arm pain score was 7.7±1.4, which significantly decreased to 0.7±1.1 at 3 days postoperatively (p<0.001). Further significant reductions were observed at 1 month (0.3±0.6, p<0.001), 6 months (0.1±0.4, p<0.001), and 12 months (0.1±0.3, p<0.001) postoperatively. According to the MacNab criteria, most patients (90.1%) exhibited excellent outcomes. The remaining patients reported good (6.9%) and fair (3.0%) outcomes, with no patient outcomes categorized as poor (Table 2). Except for transient postoperative dysesthesia reported by a minority of patients and persistent neck pain attributed to an external source distinct from the surgical site in one patient, no significant complications, such as injury to neural structures or vertebral slippage indicative of segmental instability, were observed.
Radiologic outcomes
As shown in Table 3 and Fig. 3, the radiological outcomes revealed statistically significant postoperative changes. C2-7 lordosis slightly increased from 27.7°±15.5° to 30.0°±15.1°, with a mean difference of 2.3° (p=0.005). The focal segmental angle slightly increased from 0.4°±5.0° to 1.3°±5.4°, with a mean difference of 0.9° (p=0.008), whereas the disc height slightly decreased from 5.3±1.5 mm to 5.2±1.5 mm, with a mean difference of 0.1 mm (p<0.001). The statistical significance of these three variables can be attributed to the relatively consistent differences observed in the individual measures pre-operatively and postoperatively; however, the mean differences were minor in absolute terms and were not clinically significant. Conversely, subsequent variables demonstrated changes that were both statistically significant and clinically relevant. The facet length decreased from 14.8±2.4 mm to 9.2±2.2 mm, showing a mean reduction of 5.6 mm (p<0.001). Notable increases from 3.3±1.2 mm to 7.9±2.0 mm were observed in the SAPD, with a mean difference of 4.7 mm (p<0.001), and the ID increased from 2.6±1.2 mm to 5.2±1.6 mm, with a mean difference of 2.6 mm (p<0.001). The facet resection rate was 0.4±0.1, whereas SAPD and ID expanded by 2.6±1.2-fold and 2.4±1.3-fold, respectively.
Correlation analysis between variables
The correlation between the inclinatory angle and demographic factors was investigated. The overall mean value of inclinatory angle was 18.6°±7.7° in all patients. Age groups ≤55 and >55 years did not show a significant difference in inclinatory angle, with means of 17.8°±8.3° and 19.4°±7.1°, respectively (p=0.199). Similarly, no significant difference was observed between sexes; men had a mean inclinatory angle of 19.4°±8.2°, while women had a mean inclinatory angle of 17.9°±7.1° (p=0.217). However, a significant difference was observed when comparing the affected side of the inclinatory angle; the mean for the left side was 16.5°±7.5°, while for the right side, this was 20.2°±7.4° (p=0.003). Notably, the level of operation also showed a significant correlation with inclinatory angle, with the mean progressively increasing from the C3-4 level at 12.0°±9.9° to the C7-T1 level at 21.1°±6.4° (p=0.003). These results suggest that the side of the surgical approach and spinal levels operated on are significantly correlated with variations in the inclinatory angle.
In addition, the correlations between the inclinatory angle and various radiological parameters were assessed. A moderate correlation was noted between the inclinatory angle and the facet resection rate (Pearson correlation=0.45, p<0.001), as well as the ID expansion ratio (Pearson correlation=0.3, p<0.001). No significant correlation was observed with the SAPD expansion ratio (Pearson correlation=0.1, p=0.2184). These findings indicate that the inclinatory angle demonstrates a moderate level of correlation with both the facet resection rate and ID expansion ratio. These data are presented in Table 4.
DISCUSSION
The postoperative assessment revealed a significant improvement in neck and upper arm radiating pain among the patients, with postoperative VAS scores decreasing to an average of <1 within 3 days of surgery. This enhanced condition was maintained with a slight decline over approximately 12 months. In addition, patient satisfaction was notably high, with 97% of respondents reporting favorable outcomes. Many authors have previously reported superior clinical outcomes with full endoscopy in PCF procedures [16,21,27,31,34], Similarly, BE-PCIF has been demonstrated to yield excellent clinical results for over a year or longer. Importantly, no major complications, such as iatrogenic cervical cord injury, vertebral arterial injury, dural tear, surgical site infection, or postoperative hematoma, occurred in any patient included in the study. Furthermore, all patients with postoperative transient neurapraxia exhibited only mild postoperative numbness and typically resolved within 2 days after surgery. Consequently, the safety of the BE-PCIF can also be considered an example.
Several studies have reported that resection of more than 50% of the facet joint can induce segmental instability [5,6,19]. Therefore, achieving sufficient neural decompression while preserving the original structure of the facet joint to ensure long-term postoperative stability is a technical challenge in PCF. However, cervical root decompression via the posterior approach inherently involves partial resection of the facet, thus creating an inevitable trade-off between adequate neural decompression and facet preservation in traditional PCF approaches. This issue has consistently posed a dilemma for surgeons performing posterior decompression surgery for CSR.
In contrast, PCIF modifies the approach trajectory to align more closely with the natural course of the root, minimizing unnecessary facet loss, and allowing for selective resection. This technique facilitates effective decompression of the root while preserving more of the facet joint. Radiologically, in this study, a significant postoperative increase in ID, a critical indicator for assessing complete neural decompression, was observed, with an average increase of approximately 2.6-fold, while the facet resection rate averaged 0.4±0.1. Thus, BE-PCIF achieved the crucial goal of neural decompression and preserved over half of the facet joint while maintaining stability within a safe range. Ultimately, pre- and postoperative assessments of C2-7 lordosis, focal segmental angle, and disc height showed minimal changes, confirming that BE-PCIF did not lead to postoperative segmental instability.
Upon investigating the correlation between the inclinatory angle and various demographic factors and radiological parameters, it was evident that the operated side and level, dynamic sagittal angle, facet resection rate, and ID expansion ratio exhibited statistically significant correlations. Notably, because the correlations between the dynamic sagittal angle and ID expansion ratio and inclinatory angle were weak, they were not considered clinically significant. However, two points of interest emerged : 1) a higher inclinatory angle tends to be formed when one operates at lower cervical levels and 2) an increased inclinatory angle correlates positively with a higher facet resection rate. We determined that these findings were influenced by anatomical variations in the intervertebral foramen across the cervical levels and the surgical techniques employed in BE-PCIF.
As depicted in Fig. 4, compared to the upper cervical levels, the foraminal pathway at the lower cervical levels anatomically formed a greater angle to the midline, and the lateral end of the foramen spreads further outward. Thus, when operating on the lower cervical levels compared with the upper levels, a greater inclinatory angle was necessary, and more extensive nerve decompression toward the lateral side was required. This suggests that a more substantial facet resection may be needed at lower cervical levels if adequate nerve decompression is achieved, as the foramen extends more laterally. Although less facet resection with an increasing inclinatory angle was expected in this study because skin incisions were consistently placed along the lateral margin of the spinous process in all patients, a positive correlation between a higher inclinatory angle and increased facet resections was found. Alternatively, it can be interpreted that during the process of foraminal decompression in PCIF, as the root’s silhouette is exposed and lateral decompression is performed following the nerve root pathway, it tends to naturally form the inclinatory angle in alignment with the trajectory of the root, even without deliberate intention.
When examining the correlation between the inclinatory angle and the affected side of the lesion, a trend was observed in which the inclinatory angle was significantly greater during surgery for right-sided lesions than for left-sided lesions. This phenomenon appeared to be associated with the surgeon’s technique. In this study, the surgeon was right-handed, and it was hypothesized that the differences arose from the use of a curette during neural decompression. During decompression of the distal nerve root, the surgeon carved away the bone of the SAP, covering the root using a curette, thereby expanding the ID into narrower regions. For a right-handed surgeon, rotating the curette counterclockwise facilitated bone removal, particularly on the right side, where the inherent counterforce from the inferior articular process (IAP) and SAP provides stronger resistance, simplifying bone work. Conversely, for lesions on the left, turning the curette counterclockwise caused the facet’s IAP and SAP to diverge, weakening the counterforce, and potentially resulting in insufficient bone cutting. However, it is thought that such phenomena may vary depending on the individual surgeon’s technique and the surgical tools used. Although radiological measurements indicated a lower inclinatory angle on the left side in this study, the clinical outcomes for left-sided lesions were as favorable as those on the right. A pertinent video segment is provided in Supplementary Video 1.
In recent decades, numerous surgeons have promoted advances in the surgical treatment of CSR. ACDF was first proposed in the 1950s. During the 1940s, PCF was introduced as a method to decompress nerve roots without the need for segmental fusion or instrumentation. Subsequently, both have been established as standard surgical procedures for a long period. Since the 2000s, the adoption of endoscopic techniques has rendered PCF less invasive, and the introduction of an inclined approach has further evolved the procedure into a more reasonable, safe, and efficient technique. The BE-PCIF performed in this study encompasses all the aforementioned advantages— the minimal invasiveness of endoscopy and the efficiency and safety of the inclinatory approach—making it a highly advanced technique in the contemporary surgical landscape.
The limitations of this study include its single-center design and retrospective nature. Thus, future comparative studies using other surgical techniques or prospective studies are required to validate these findings.
CONCLUSION
BE-PCIF is an effective surgical method for alleviating pain and improving clinical outcomes in patients with CSR. Use of the inclinatory angle approach has been shown to effectively expand the cervical foraminal canal, facilitating sufficient neural decompression. Furthermore, achieving adequate decompression at lower cervical levels requires higher inclinatory angles.
PDF Links
PubReader
ePub Link
XML Download
Full text via DOI
Download Citation 
Supplement
Print 
