Relationship Between C2 Semispinalis Cervicis Preservation and C2 Spinous Process Morphology During Cervical Laminoplasty Involving C3

Abstract

Study Design

Retrospective study

Objectives

Due to anatomical variations in the semispinalis cervicis insertion in the C2 spinous process, complete preservation is not always possible when the C3 level is included in a cervical laminoplasty. Three-dimensional computed tomography was used to evaluate the relationship between the incidence of semispinalis cervicis injury and the C2 inter-spinous angle.

Methods

We included 95 patients who underwent a cervical laminoplasty that included a C3 laminectomy for cervical myelopathy. Patients with a C2 inter-spinous angle above and below 60° were classified into wide- and narrow-angled groups, respectively (n = 48 and n = 47). Whether the C2 semispinalis cervicis insertion was preserved, or detached and reattached was reviewed from surgical records. The pre and postoperative C2-C7 lordosis and range of motion (ROM) were measured, and clinical outcomes were obtained from the patient charts.

Results

The C2 semispinalis cervicis was preserved in 47 patients (97.9%) in the wide-angled group but only in 14 patients (29.8%) in the narrow-angled group (P < .001). The postoperative C2-C7 lordosis extension and ROM were significantly greater in the wide-angled (P = .048 and .036). Postoperative neck pain was significantly greater in the narrow-angled (P = .018).

Conclusions

The morphology of the C2 spinous process indicates that a C2 semispinalis cervicis insertion preservation is possible during a cervical laminoplasty that includes a C3 laminectomy. A careful surgical procedure should be conducted when the C2 inter-spinous angle is above 60° to increase the likelihood of achieving this preservation and thereby obtaining a more favorable clinical outcomes.

Introduction

A cervical laminoplasty is a widely used surgical technique for cervical spondylotic myelopathy (CSM) caused by cervical spondylosis or ossification of the posterior longitudinal ligament (OPLL).^1-3 Despite many reports of a neurological improvement after a laminoplasty, there are several notable postoperative complications of this procedure, including a reduced range of motion (ROM), loss of cervical lordosis, and axial neck pain.^4-6 These complications may arise due to extensive dissection of the muscles attached to the spinous process and lamina and/or injury to the semispinalis cervicis muscle during conventional cervical laminoplasty procedures.^7,8 Various modified laminoplasty techniques have been described that aim to resolve this problem, but this places a greater technical demand on the surgeon compared with conventional unilateral open-door laminoplasty.^11-13

Several prior studies have reported that the semispinalis cervicis plays an important role in cervical dynamic stability, particularly the insertion into the axis (C2).^8-10 These same reports have also emphasized the importance of minimizing any detachment of the C2 semispinalis cervicis during a conventional laminoplasty.^8-10 However, when the C3 is included in a conventional unilateral open-door laminoplasty procedure, the C2 semispinalis cervicis is difficult to preserve for the following reasons: 1) the course of the C2 semispinalis cervicis (originating from the T1 transverse process insertion to the C2 spinous process); 2) the tensile force applied to the semispinalis cervicis when the patient is placed in the intraoperative kyphotic position for better visualization, which forces its migration to the distal side after injury; and 3) the vulnerability of the semispinalis cervicis muscle fiber.

Hence, when decompression at the C3 level is needed, it would be prudent to identify patients in whom the semispinalis cervicis can be preserved and to conduct the surgery with care to minimize the possibility of damage. This study aimed to provide useful information for predicting the likely preservation of the semispinalis cervicis muscle using three-dimensional (3D) computed tomography (CT) evaluations prior to surgery.

Material and Methods

Study Design

The study cohort compromised 95 patients who underwent a cervical laminoplasty, that included a C3 laminectomy, for multi-level CSM with C3 cord compression, between August 2011 and December 2018. These cases were retrospectively reviewed. The study protocol was approved by the institutional review board (IRB) of Asan Medical Center (IRB no. A20210320). Due to the retrospective nature of the analyses and the use of anonymized patient data, the requirements for patient-informed consent to participate was waived by IRB.

The study exclusion criteria were as follows: 1) a previous record of C3 laminoplasty surgery; 2) a previous record of simultaneous anterior or fusion surgery performed; 3) a previous record of dome-shaped C3 partial laminectomy surgery; 4) preoperative ROM ≤ 5° due to diffuse idiopathic skeletal hyperostosis or OPLL; 5) preoperative neck pain > 3 according to a visual analog scale (VAS) score; 6) k-line negative patients; 7) patients with segmental instability; 8) patients with no postoperative follow-up ≥ 2 years; and 9) the presence of other disease entities, including tumors, traumas, and infections.

Vertebral artery anomalies were screened before surgery in all CSM patients at our institution using routine 3D-CT angiography. We hypothesized that because the C2 semispinalis cervicis is attached to the C2 spinous process, the anatomy of the semispinalis cervicis could be predicted according to the C2 spinous process morphology via 3D-CT. We divided the study patients into 2 groups in accordance with the C2 inter-spinous angle. Patients with a C2 inter-spinous angle > 60° were placed into a wide-angled group and those with a C2 inter-spinous angle < 60° into a narrow-angled group (Figures 1 and 2).

Figure 1.

Two patients’ semispinalis cervicis muscle morphology prediction. (A) Group N patient’s preoperative 3D-CT (C2ISA 31 degree) ; intraoperative C2 SSC reattachment procedure after detachment (B) Group W patient’s preoperative 3D-CT (C2ISA 65 degree) ; intraoperative C2 SSC preservation.

Figure 2.

Illustration of C2 spinous process and semispinalis muscle anatomy in two groups (A) Group N (B) Group W.

Operative Technique and Postoperative Management

All surgical procedures were performed by a single spine surgeon (DHL). Briefly, a posterior midline incision was performed after preoperatively verifying the level using C-arm fluoroscopy. This approach was then extended to the C2 spinous process level to visualize the C3 lamina-facet junction. At the C2 level, the approach was carefully performed from the medial to the lateral side of the semispinalis cervicis at the distal level of the spinous process to enable preservation of the semispinalis cervicis. When detachment was required, approximately one-third of the medial at both semispinalis cervicis insertion sites was first detached, and visualization of the C3 lamina for laminectomy was then achieved using a Gelpi retractor. If sufficient visualization could not be achieved, detachment of more than one-third of the semispinalis cervicis was performed, and a unilateral open-door cervical laminoplasty, including a C3 laminectomy, was conducted as the first-line treatment. Unilateral open-door laminoplasties were performed using a plate and two screws in the lamina and lateral mass, respectively. Approximation of the cutting margin in the semispinalis cervicis to the periosteum was conducted when the detachment was performed for less than one-third. If more than one-third of the semispinalis cervicis was detached, a hole was made medial to lateral using a towel clip or burr on the lateral side of the spinous process. The detached semispinalis cervicis was then fixed onto the hole using a double row suture while maintaining the appropriate tension. In addition, when decompression was required at the C2 and C7 levels, dome-shaped laminoplasty was performed to preserve the muscular structure by tilting the microscope toward the distal side of C2 or the proximal side of C7. When partial decompression would suffice, a dome-shaped partial laminectomy was also performed in which only the proximal side below the C4 level was decompressed. All patients were encouraged to wear cervical neck collars for 1 month after surgery, and underwent supervised rehabilitation for neck strengthening and stretching.

Radiological Analysis

Preoperative 3D-CT angiography was used to evaluate the specific anatomy around the C2 spinous process and the T1 transverse process. The distances between the C2 spinous processes, T1 transverse processes, C2 inter-spinous angle, and semispinalis cervicis muscle angle were measured (Figure 3). The height of the C2 spinous process was measured using the sagittal view to calculate the maximum distance from the lamina. The C2 inter-spinous diameter was measured as the distance between the lateral margins of the C2 spinous process, and the T1 inter-transverse process diameter was measured as the distance between the lateral margins of the T1 transverse process. The C2 inter-spinous angle was measured as the angle between two lines inside the spinous process, and the semispinalis cervicis muscle angle as the angle between the lateral ends of the C2 spinous process and lines connecting both lateral ends of the T1 transverse process. C2-7 lordosis at flexion and extension was evaluated preoperatively and 2 years postoperatively using a simple radiograph. The difference was defined as the ROM.

Figure 3.

Radiologic parameters around C2 semispinalis cervicis muscle (A) C2ISA (C2 inter-spinous process angle) (B) C2ISD (C2 inter-spinous process distance), T1ITD (T1 inter-transverse process distance) (C) SMA (Semispinalis muscle angle).

Clinical and Functional Scores

Clinical and functional scores were measured using the patients’ records to determine the VAS for neck pain (VAS-N), and Annex 1 Korean Neck Disability Index both preoperatively and at 2 years postoperatively.

Statistical Analysis

The paired t-test, chi-squared test, and Fisher’s exact test were used to compare differences between each parameter analyzed and P values < .05 were considered to indicate statistical significance. All statistical analyses were performed using SPSS version 21.0 (IBM Corporation, Armonk, NY, USA).

Results

Demographic Factors

The mean (± standard deviation) age of the included study patients was 66.8 ± 11.5 years and this cohort comprised 72 men and 23 women. Underlying diseases included OPLL (n = 67), spondylosis (n = 26), and ossification of the yellow ligament (OYL) (n = 2). The C2 semispinalis cervicis was preserved in 61 patients (64.2%) but required reattachment after detachment in 34 patients (35.8%). Notably this preservation was achieved in 47/48 patients (97.9%) in the wide-angled group but was possible only in 14/47 patients (29.8%) in the narrow-angled group (P < .001). The mean lesion level was 4.4 ± .8. Demographic information for the study subjects is summarized in Table 1.

Table 1.

Comparison of Patient Demographics Between the two Study Groups.

	Total	Group W (n = 48)	Group N (n = 47	P value
Age (yr)	66.8 ± 11.5	68.4 ± 10.5	65.3 ± 12.3	.189
Sex				.766
- Male (%)	72 (75.8%)	37 (77.1%)	35 (74.5%)
- Female (%)	23 (24.2%)	11 (22.9%)	12 (25.5%)
Associated disease				.414
- OPLL (%)	67 (70.5%)	32 (66.7%)	35 (74.5%)
- Spondylosis (%)	26 (27.4%)	15 (31.3%)	11 (23.4%)
- OYL (%)	2 (2.1%)	1 (2.1%)	1 (2.1%)
C2 SSC				< .001
- Preservation	61 (64.2%)	47 (97.9%)	14 (29.8%)
- Reattachment	34 (35.8%)	1 (2.1%)	33 (70.2%)
Decompression level	4.4 ± 0.8	4.4 ± 0.8	4.3 ± 0.9	.675

* The data shown is the mean values with the standard deviation for continuous variables and number of cases for categorical variables.

Group W, wide-angled group; Group N, narrow-angled group.

OPLL, ossification of the posterior longitudinal ligament; OYL, ossification of the yellow ligament; SSC, semispinalis cervicis.

Radiological Factors

Radiological results revealed that the mean C2 inter-spinous angle was 67.7 ± 7.5° in the wide-angled group and 36.3 ± 9.6° in the narrow-angled group. The C2 inter-spinous diameter was 19.0 ± 3.1 mm in the wide-angled group and 12.5 ± 2.9 mm in the narrow-angled group (P < .001). The semispinalis cervicis muscle angle was 32.0 ± 3.7° in the wide-angled group, and 34.8 ± 4.0° in the narrow-angled group, which was also a statistically significant difference (P = .001). These and other results are summarized in Table 2.

Table 2.

Comparison of Radiological Factors Between the two Study Groups.

	Total	Group W (n = 48)	Group N (n = 47)	P value
C2ISA (°)	52.2 ± 17.9	67.7 ± 7.5	36.3 ± 9.6	< .001
C2ISD (mm)	15.3 ± 4.0	18.0 ± 3.1	12.5 ± 2.9	< .001
T1ITD (mm)	63.1 ± 4.3	63.3 ± 3.7	62.9 ± 5.0	.691
SMA (°)	33.4 ± 4.1	32.0 ± 3.7	34.8 ± 4.0	.001
C2SH (mm)	22.1 ± 3.2	22.4 ± 3.5	21.9 ± 2.9	.466
Preop. C2-7 lordosis (°)
- Neutral position	11.2 ± 7.9	10.6 ± 7.5	11.8 ± 8.5	.482
- Flexion position	−13.4 ± 11.4	−14.7 ± 9.7	−12.1 ± 12.8	.300
- Extension position	22.9 ± 9.8	21.0 ± 8.3	24.9 ± 10.8	.063
Postop. C2-7 lordosis (°)
- Neutral position	6.4 ± 8.7	8.0 ± 6.6	4.7 ± 10.2	.061
- Flexion position	−14.3 ± 11.7	−14.7 ± 9.0	−13.9 ± 13.9	.742
- Extension position	16.0 ± 10.4	17.3 ± 8.1	13.7 ± 10.9	.048
Preop. ROM (°)	36.2 ± 13.0	35.7 ± 9.9	36.8 ± 15.5	.691
Postop. ROM (°)	30.3 ± 12.3	31.9 ± 11.2	28.8 ± 13.2	.036

* The data shown is the mean values with the standard deviation.

Group W, wide-angled group; Group N, narrow-angled group.

C2ISA, C2 inter-spinous angle; C2ISD, C2 inter-spinous process distance.

T1ITD, T1 inter-transverse process distance; SMA, semispinalis cervicis muscle angle.

C2SH, C2 spinous process height; Preop., preoperative; Postop., postoperative.

ROM, range of motion.

Clinical Factors

The mean postoperative neck pain score (VAS-N) was significantly greater in the narrow-angled group (2.5 ± 2.6 vs 1.6 ± 1.9; P = .018). The postoperative C2-C7 lordosis in the extension position was significantly lower in the narrow-angled group (14.7 ± 10.9 vs 17.3 ± 8.1; P = .048) (Figure 4). The postoperative ROM was also lower in the narrow-angled group (28.8 ± 13.2 vs 31.9 ± 11.2; P = .037). These results are presented in Table 3.

Figure 4.

Two patients who underwent cervical laminoplasty including C3 laminectomy (A,B) Group N patient preoperative and postoperative lateral radiograph in extension (each C2-C7 lordosis value: 26, -3 degree) (C,D) Group W patient preoperative and postoperative lateral radiograph in extension (each C2-C7 lordosis value: 30, 28 degree).

Table 3.

Comparison of Clinical Factors Between the two Study Groups.

	Total	Group W (n = 48)	Group N (n = 47)	P value
Preoperative
- VAS-N	2.1 ± 2.0	2.5 ± 2.4	1.8 ± 1.6	.157
- AKNDI	27.9 ± 17.2	28.0 ± 15.1	27.8 ± 19.1	.967
Postoperative 2-year
- VAS-N	2.1 ± 2.3	1.6 ± 1.9	2.5 ± 2.6	.018
- AKNDI	21.2 ± 17.2	20.9 ± 15.1	21.6 ± 19.5	.875

* The data shown is the mean values with the standard deviation.

Group W, wide-angled group; Group N, narrow-angled group.

VAS-N, visual analog scale-neck; AKNDI, Annex 1 Korean Neck Disability Index.

Discussion

Compared with previous laminectomy and fusion methods, a cervical laminoplasty is currently believed to be the most appropriate surgical technique to preserve cervical motion in patients with myelopathy due to cervical degenerative spinal disease. A laminoplasty can also be used as an indirect decompression method for multi-level cord compression and has the advantage of a posterior approach that is familiar to spine surgeons.^1-3 Notably however, several studies have reported that posterior ligament and muscle damage from these surgeries can cause axial neck pain due to a postoperative kyphotic deformity and ROM limitation.^4-6

Conley et al. confirmed that the semispinalis cervicis is the main extensor muscle of the cervical spine by investigating its cross-sectional area and the intensity of the posterior muscle for neck motion. ¹⁵ Vasabada et al. reported that the semispinalis cervicis plays a major role in neck extension moment-generating capacity using 3D biomechanical studies. ¹⁶ Schomacher et al. reported that the semispinalis cervicis is also active predominately in extension. ¹⁷ Several other studies reported that C7 nuchal ligament injury could also be a major cause of axial neck pain.^18-20

For the C7 nuchal ligament, a C7 partial laminectomy (i.e.,, dome-like laminectomy) is most commonly used, and a method has been established, to some extent, for the preservation of the nuchal ligament. ²¹ However, in the case of the C2 semispinalis cervicis, it is practically quite difficult to preserve the muscle while performing the procedure at the C3 level due to the anatomy of this structure. Despite the development of various laminoplasty methods, such as a muscle preserving laminoplasty to resolve this problem, it is a more complex procedure than a conventional unilateral open-door laminoplasty at the C3 level.^12-14 Furthermore, high tension is applied to the tendinous portion in the intraoperative neck flexion position during surgery, which causes the C2 semispinalis cervicis to be highly vulnerable to thermal injury in the operative field. In a prior study by Lee et al., preservation of the C2 semispinalis cervicis was achieved in only 67% of patients after conventional laminoplasty. ¹³ Comparably, only 64.2% of our current study cases showed preservation of the C2 semispinalis cervicis muscle.

Takeuchi et al. reported anatomical variations in the width and height of the semispinalis cervicis, and in the length and width of the C2 spinous process, in 24 cadavers. ²² We thus postulate that careful preoperative anatomical assessment of the semispinalis cervicis can provide very useful information for predicting the likely preservation or damage of the semispinalis cervicis during a laminoplasty. We speculated that when a kyphosis progression was predicted in patients in whom preservation of the C2 semispinalis cervicis had failed, an alternative surgical approach such as a muscle preserving laminoplasty could potentially prevent this failure. Our current observations indicated that the semispinalis cervicis preservation was successful in most patients with a C2 inter-spinous angle > 60°, but far less likely in cases with a lower angle than this.

Iizuka et al. reported a postoperative interlaminar bony fusion incidence of 53% at C2-3 after laminoplasty. ²³ Lee et al. reported that a significantly improved cervical ROM after a C3 laminectomy was achieved by preventing adjacent interlaminar bony fusion when cord compression was present at the C3 level. ²⁴ Takeuchi et al. reported in their prior report that the C2–7 ROM was improved in patients who underwent a C3 laminectomy and C4–7 laminoplasty vs a C3–7 laminoplasty. ¹⁰ These authors further reported that there was also less reduction in the ROM during neck extension in the former cases. We also believe that performing laminectomy instead of a laminoplasty at the C3 level during a unilateral open-door laminoplasty would lead to an improved preservation of the semispinalis cervicis by reducing the space exposed for surgery. For this reason, the C3 level has recently mainly undergone a laminectomy instead of laminoplasty in patients with cervical myelopathy, and we thus focused on C3 laminectomy cases in our present analyses. We found from our present analyses that both C2 the inter-spinous angle and C2 inter-spinous diameter differed between our two study groups. In addition to the C2 inter-spinous angles, the C2 inter-spinous distances were also greater in the patients with wide angles. However, we contend that the C2 inter-spinous angle is a more important factor than the C2 inter-spinous diameter for preserving the semispinalis cervicis muscle because it is minimally affected by the patient’s sex or body size or by the size of the cervical spine.

If preoperative 3D-CT images are difficult to obtain routinely, the inter-spinous angle of the C2 spinous process can be measured using the anterior to posterior (AP) route on a radiograph. It must be noted however that if these radiographs are not true AP images, the correct angle may not be obtained and accurate measurements may in fact be difficult to obtain because the C2 spinous process does not look symmetrical in the 2D plane for every patient.

The present study had some limitations of note. First, not all of the laminoplasties in our study patients were performed at the same cervical level. Nevertheless, there were no statistical differences found between the two groups. Second, the quality of our evidence was limited by the retrospective nature of the analyses. Finally, regardless of the shape of the C2 spinous process, there can be differences in the preservation of the semispinalis cervicis due to differences in interpreting the anatomy of the muscle, and this is dependent on the surgeon’s technical skills.

Conclusion

In conclusion, predicting the likelihood of a C2 semispinalis cervicis preservation through a preoperative CT evaluation can assist with determining the most appropriate surgical approach. Of note in particular, when the C2 inter-spinous angle is above 60°, a careful surgical procedure greatly increases the chance that the semispinalis cervicis can be preserved to obtain more favorable clinical outcomes.

ORCID iDs

Dong-Ho Lee https://orcid.org/0000-0003-3704-6355

Sehan Park https://orcid.org/0000-0001-8959-8579

Jae Hwan Cho https://orcid.org/0000-0002-1178-9778