Current affiliation : Graduate School of Medical Science and Engineering, Korea Advanced Institute of Science and Technology (KAIST), Daejeon, Korea
To compare the anterior cervical discectomy and fusion (ACDF) and posterior cervical fusion (PCF) with wide facetectomy in the treatment of parallel-shaped bony foraminal stenosis (FS).
Thirty-six patients underwent surgery due to one-or-two levels of parallel-shaped cervical FS. ACDF was performed in 16 patients, and PCF using CPS was performed in 20 patients. All patients were followed up at 1, 3, 6, and 12 months postoperatively. Standardized outcome measures such as Numeric rating scale (NRS) score for arm/neck pain and Neck disability index (NDI) were evaluated. Cervical radiographs were used to compare the C2–7 Cobb’s angle, segmental angle, and fusion rates.
There was an improvement in NRS scores after both approaches for radicular arm pain (mean change -6.78 vs. -8.14,
In the treatment of parallel-shaped bony FS up to two surgical levels, segmental angle improvement was more favorable in patients who underwent ACDF. However, PCF with wide facetectomy using CPS should be considered as an alternative treatment option in cases where the anterior approach is burdensome.
Although a cervical posterior foraminotomy without fusion is a good surgical option for treatment of mild or moderate cervical foraminal stenosis (FS), it is not appropriate for the treatment of severe bony FS when the nerve root is compressed across the entire path of the cervical foramen [
Since wide facetectomy for parallel-shaped FS is not possible with the lateral mass screw, we did PCF with cervical pedicle screw (CPS). Though there has been some controversy over the use of CPS in degenerative disease, CPS was used for sufficient facetectomy and foraminal decompression.
We previously demonstrated the effectiveness and safety of CPS placement even in patients with degenerative cervical spinal pathology such as parallel-shaped cervical FS [
The study protocol was approved by the Institutional Review Board of Asan Medical Center (Approval No. AMC IRB 2020-0700). We retrospectively reviewed medical records of patients who underwent cervical spinal surgery by ACDF or PCF to treat one-or-two levels of parallel-shaped FS between March 2012 and August 2016. The selection of surgical approach depended on the surgeons’ preferences. No patients with cervical myelopathy or cervical soft disc herniation were included in this study. Only patients with radiculopathy caused by bony hypertrophy through the cervical foramen who did not respond to non-operative treatment for more than two months were included. Especially, the shape of narrowed foramen on axial CT scan was seen only as a parallelshape, not V-shape, was included, and this was regarded as the narrowing through an entire path of the foramen (
We included patients who were follow-up for at least 12 months. We excluded patients with serious cardiovascular, pulmonary, or cerebrovascular comorbidity, or history of anticoagulant treatment (i.e., warfarin or clopidogrel). All patients were underwent preoperative CT scans, magnetic resonance imaging, and X-ray imaging.
For an ACDF, the standard Smith-Robinson approach was used. After level identification with X-ray examination, discectomy was performed. To achieve a wide foraminal decompression, the uncinate process removal of the affected site was performed. Properly sized interbody spacers made of iliac bone or allograft block were inserted, followed by anterior plate fixation [
For PCF procedures, freehand CPS insertion was primarily considered if the outer diameter of the cervical pedicle was greater than 3.0 mm on axial CT images. Patients were placed prone and Gardner-Wells tongs were applied with 3 or 5 pounds of traction to achieve a maximally horizontal head position. Motor evoked potential monitoring was used throughout the procedure. The entry point of the screw was determined by the notch level in the sagittal plane, medial to the lateral border of the superior articular process by one-quarter of its width in the axial plane. A small pilot hole was made at the predetermined entry point with a 1.8-mm diameter match head-type burr. After forming a trajectory through the cancellous channel of the pedicle and tapping with a 3.5-mm diameter tap, the screw was inserted. If we were not sure of the safety of the pedicle trajectory after repeated probing with a ball tip probe, we converted to lateral mass screw placement. A more detailed technical description of our technique was described previously [
After CPS placement, a wide facetectomy was performed for decompression of the entire path of the nerve root. After cutting of inferior articular process with an osteotome, drilling of the superior articular process was done with a 3.0 mm sized match head type burr and 1 or 2 mm Kerrison rongeur. We considered complete decompressin to be achieved after complete removal of the tip of the superior articular process and identification of the lateral dura margin and origin of the nerve root (
All patients were advised to wear a soft collar for four weeks after the operation and were followed up at 1, 3, 6, and 12 months. We compared clinical outcomes between the two groups, including any change in the Numeric rating scale (NRS) score for radiating arm pain and neck pain, complications, surgical time, estimated blood loss (EBL), pre and post operative white blood cell (WBC) count, and in the Neck disability index (NDI) at the 12 months follow-up were used in this study.
We reviewed a series of lateral cervical X-rays of patients in an upright, neutral position and measured the Cobb angle between the lower border of the C2 body and the lower border of the C7 body to evaluate cervical sagittal alignment (
We used SPSS ver. 18.0 statistical software package (SPSS Inc., Chicago, IL, USA) for statistical analysis. This study presented comparing data from 16 and 20 subjects (10≤ n ≤30). The normality of the distributions was assessed using the Shapiro-Wilk test. Groups were compared for demographic data (mean age, follow-up duration), clinical outcomes (surgical time, WBC count), and radiologic outcomes of surgery by student t-test after the normality test (
Sixteen patients underwent ACDF, and 20 patients underwent PCF using CPS. There were no significant differences between the two groups in terms of sex, age, follow-up duration, operation level, and surgical level. In both groups, surgery for the C5–7 level was the most common (Ant, 43.8%; Post, 55%), and two-level surgery was more common than one-level surgery (75% vs. 25%). Patient characteristics are described in
Preoperative baseline and postoperative outcomes are presented in
Radiologic outcomes are presented in
We compared the outcomes of ACDF with PCF using CPS in a very selective and homogenous group of patients with parallel-shaped FS. Although decompressive surgery such as a posterior cervical foraminotomy is approprate for common cervical FS, it is not appropriate for treatment of an FS with advanced stage degeneration where the entire path of the foramen is narrowed from the entrance zone of the nerve root of the spinal canal to the exit zone. In this situation, ACDF with uncinate process removal or PCF with wide foraminal decompression through facet resection is recommended to achieve sufficient decompression [
We previously reported that parallel shaped bony FS is an indication of ACDF or posterior fusion surgery with aggressive foraminotomy [
ACDF remains the gold standard for treating cervical spondylotic radiculopathy and FS due to discs and spurs at one or two levels [
When using PCF for parallel shaped FS, it is necessary to remove the entire facet joint to achieve sufficient decompression to the lateral side of the foramen, which inevitably requires pedicle fixation. This wide decompression is similar to Ames’ grade 2 osteotomy (complete facet joint/Ponte osteotomy) for deformity correction [
In addition, the fusion bed around the lateral mass, which is an important bone graft recipient site, may be smaller because of the area occupied by the head of lateral mass screw [
The biomechanical advantage of the CPS over the lateral mass screw has been previously established. Despite these biomechanical benefits, the risk of possible neurovascular complications have been identified as areas of concern [
The outcomes of radiating arm pain were comparable between groups, and there was no difference in overall NDI scores and NRS neck pain scores at last follow-up. Although there were no differences in clinical outcomes overall between the two treatment methods, neck pain was significantly higher in patients who underwent PCF until 6 months after surgery. EBL of the PCF group was also significantly higher than the ACDF group, and it is an indirect indicator that PCF is more invasive than ACDF. This finding highlights a major disadvantage of conventional midline posterior surgical approaches for cervical fusion surgery as extensive subperiosteal dissection and detachment of paraspinal muscles from the lamina can lead to postoperative cervical muscular pain and spasming [
The C2–7 Cobb’s angle for the evaluation of cervical alignment showed no difference in global cervical lordosis between the two groups. However, the ACDF group had a larger mean segmental angle immediately after surgery, and this difference between was maintained statistically significant at the last follow-up. This gap indicates that ACDF, which can support the anterior column by cage insertion, is more favorable for segmental angle correction and maintenance than PCF. As has previously been demonstrated in biomechanical studies, the presence of interbody cages significantly improves overall construct stiffness when compared with posterior-only implant systems [
Our study had several limitations. First, it was a retrospective observational study involving a small number of patients. Second, the severity of baseline neck pain scores differed between patients who received ACDF or PCF, suggesting a difference in the degree of spondylosis. Typically, the surgical indications of ACDF and PCF are not the same. In order to compare the two surgical methods objectively, our study needs a clear distinction for surgical indications. Third, postoperative 12-month CT scans were not performed on all patients included in this study to assess bone fusion. Last, although complemented by a linear mixed model, there was a loss of the interval values of some patients during the whole follow-up period repeated measurements.
ACDF with uncinate process removal is better than PCF in treating parallel shaped bony FS up to two surgical levels in terms of maintaining larger segmental angles and less postoperative neck pain. However, in situations with severe cervical spondylosis with deformity or where an anterior approach is difficult, PCF with wide facetectomy using CPS may be one alternative treatment option for parallel shaped FS.
No potential conflict of interest relevant to this article was reported.
This type of study does not require informed consent.
Conceptualization : SWR, JHP
Data curation : SL, HC
Formal analysis : SL, IC
Funding acquisition : SWR, JHP
Methodology : SWR, IC
Project administration : DCC, JHP
Visualization : DCC, HC
Writing - original draft : SL, IC
Writing - review & editing : SL, DCC, IC, JHP
The shape of narrowed foramen on axial computed tomography scans. A : Parallel-shaped foraminal stenosis (FS) on the intervertebral level. The entire path of foramen was narrowed from the entrance zone to the exit zone. B : In contrast to parallel-shaped FS, V-shaped FS becomes wider at the exit zone.
A 70-year-old man with tingling in both arms and left elbow flexion weakness. Severe parallel shape foraminal stenosis (FS) was seen on preoperative computed tomography (cT) imaging, with the left FS being more pronounced. A : Preoperative sagittal cT image show the c5–6, c6–7 foramen. B : Preoperative axial cT image of the c5–6 level foramen. c : Sagittal cT image of 12 months after c5–6 and c6–7 wide facetectomy (Ponte osteotomy) with fusion. cervical pedical screws were inserted into both sides of c6 and c7, and lateral mass screws were inserted into both sides of the c5. d : Axial cT image of the c5–6 level 12 months after surgery. Both sides of the c6 superior articular processes were removed, and the entire path of c5–6 foramen was widened.
A lateral cervical X-ray of a patient in a neutral position shows the c2–7 cobb angle measurement between the lower border of the c2 body and the lower border of the c7 body. The segmental angle measurement is made between the lower border of the distal instrumented vertebral body and the upper border of the proximal instrumented vertebral body. A : Preoperative image. B : Immediate postoperative image. The segmental angle of surgical index level increased after anterior cervical decompression, c5–6–7. c : X-ray image of 12 months after surgery. The segmental angle and c2–7 cobb angle remained well after 12 months.
An illustration showing an exposed screw tip or thread of a lateral mass screw during resection of the superior and inferior articular process. In cases of wide facetectomy (red dashed line), the lateral mass screw is partially exposed, but the trajectory of the cervical pedicle screw is not affected by it.
Characteristics of patients who underwent AcdF with unicinate removal or PcF with wide facetectomy for parallel-shaped bony FS
ACDF (n=16) | PCF (n=20) | ||
---|---|---|---|
Sex, male : female | 11 : 5 | 19 : 1 | 0.069 |
Age (years) | 52.8±11.2 | 58.7±9.5 | 0.096 |
Follow up duration (months) | 22.4±11.6 | 20.8±9.8 | 0.649 |
Operative level | 0.459 | ||
C3–5 | 2 (12.5) | 1 (5.0) | |
C4–6 | 3 (18.8) | 3 (15.0) | |
C5–6 | 3 (18.8) | 1 (5.0) | |
C5–7 | 7 (43.8) | 11 (55.0) | |
C6–7 | 1 (6.3) | 4 (20.0) | |
Number of surgical level | 0.652 | ||
1 level | 4 (25.0) | 5 (25.0) | |
2 level | 12 (75.0) | 15 (75.0) | |
Comorbidity | 2 (12.5) | 1 (5.3) | 0.582 |
Lou Gehrig's disease 1 | Stroke 1 | ||
T-spine metastatis 1 |
Values are presented as mean±standard deviation or number (%). ACDF : anterior cervical discectomy and fusion, PCF : posterior cervical fusion, FS : foraminal stenosis
Clinical outcomes in patients
ACDF (n=16) | PCF (n=20) | ||
---|---|---|---|
Mean NRS of arm pain | |||
Pre-operation | 8.2±0.8 | 8.6±1.0 | 0.208 |
Postoperative 1 M | 1.2±0.7 | 1.2±1.9 | 0.967 |
Postoperative 3 M | 0.9±0.9 | 0.8±1.3 | 0.905 |
Postoperative 6 M | 1.0±1.4 | 0.5±1.0 | 0.341 |
Postoperative 12 M | 0.6±1.2 | 0.4±0.8 | 0.651 |
Last follow-up | 1.4±1.2 | 0.5±1.1 | 0.100 |
Mean NRS of neck pain | |||
Pre-operation | 1.7±2.3 | 4.4±3.3 | 0.010 |
Postoperative 1 M | 0.8±0.9 | 3.2±2.7 | 0.004 |
Postoperative 3 M | 0.5±1.1 | 1.7±1.7 | 0.026 |
Postoperative 6 M | 0.2±0.4 | 1.2±1.3 | 0.028 |
Postoperative 12 M | 0.5±1.2 | 0.4±0.8 | 0.910 |
Last follow-up | 0.0±0.0 | 0.5±1.5 | 0.179 |
Complications | |||
Any complication, yes | 2 (12.5) | 1 (5.0) | 0.660 |
Dysphagia 1 | C5 palsy 1 | ||
Myelitis 1 | |||
Dysphagia, yes | 1 (6.3) | 0 (0.0) | 0.457 |
Invasiveness | |||
Surgical time (minutes) | 162.44±42.34 | 186.75±41.33 | 0.092 |
EBL (mL) | 91.25±62.91 | 420.00±123.97 | <0.001 |
WBC pre-operation (/μL) | 6962.5±1831.9 | 7035.0±1956.7 | 0.910 |
Postoperative 1 day | 9693.8±2841.0 | 9325.0±2693.6 | 0.693 |
Post-pre change | 2731.3±2200.5 | 2290.0±1957.4 | |
Post-pre paired t-test |
Values are presented as mean±standard deviation or number (%). ACDF : anterior cervical discectomy and fusion, PCF : posterior cervical fusion, NRS : Numeric rating scale, M : months, EBL : estimated blood loss, WBC : white blood cell
Radiologic outcomes
ACDF (n=16) | PCF (n=20) | ||
---|---|---|---|
Interspinous distance, on the last dynamic X-ray (mm) | 0.9±0.6 | 0.4±0.2 | 0.006 |
Fusion bridge formation, in case of postoperative 12 M CT acquired | 13/13 (100.0) | 13/13 (100.0) | 0.693 |
C2–7 Cobb angle (°) | |||
Pre-operation | 14.1±11.9 | 12.7±9.9 | 0.687 |
Postoperative 1 M | 15.8±10.2 | 12.9±7.8 | 0.430 |
Postoperative 3 M | 15.9±9.6 | 15.8±10.1 | 0.959 |
Postoperative 6 M | 13.5±8.6 | 12.4±8.8 | 0.778 |
Postoperative 12 M | 15.1±8.4 | 14.0±7.3 | 0.758 |
Last follow-up | 16.2±7.9 | 14.8±8.5 | 0.691 |
Segmental angle (°) | |||
Pre-operation | 3.1±5.6 | 2.9±6.0 | 0.955 |
Postoperative 1 M | 8.1±6.9 | 4.4±4.6 | 0.126 |
Postoperative 3 M | 8.4±7.2 | 5.9±5.8 | 0.303 |
Postoperative 6 M | 7.9±7.3 | 5.2±4.6 | 0.289 |
Postoperative 12 M | 7.9±4.9 | 6.4±6.1 | 0.536 |
Last follow-up | 9.4±2.7 | 3.3±5.1 | 0.004 |
Values are presented as mean±standard deviation or number (%). ACDF : anterior cervical discectomy and fusion, PCF : posterior cervical fusion, M : months, CT : computed tomography
Linear mixed model showing the difference between groups according to the change of time
Change |
F/U months | ACDF (n=16) | Time |
PCF (n=20) | Time |
Time×group |
Group |
---|---|---|---|---|---|---|---|
C2–7 Cobb angle (°) | 1 | 16 (1.63±10.92) | 0.916 | 12 (-0.42±6.60) | 0.120 | 0.302 | 0.672 |
3 | 15 (1.33±11.15) | 16 (2.00±6.29) | 0.949 | ||||
6 | 11 (0.91±11.96) | 12 (2.92±5.07) | 0.513 | ||||
12 | 13 (0.23±8.75) | 9 (1.56±7.43) | 0.581 | ||||
Last | 9 (1.67±9.43) | 14 (5.57±4.54) | 0.142 | ||||
Segment angle (°) | 1 | 16 (5.00±8.05) | 0.974 | 12 (1.50±5.95) | 0.793 | 0.848 | 0.226 |
3 | 15 (5.33±8.71) | 16 (2.19±5.39) | 0.188 | ||||
6 | 11 (6.09±9.27) | 12 (2.83±5.70) | 0.300 | ||||
12 | 13 (4.62±6.46) | 9 (1.78±6.30) | 0.218 | ||||
Last | 9 (6.33±6.02) | 14 (1.43±4.93) | 0.292 | ||||
Arm pain Score, NRS | 1 | 16 (-7.00±0.97) | 0.340 | 18 (-7.56±2.36) | 0.615 | 0.467 | 0.296 |
3 | 15 (-7.27±1.10) | 17 (-7.76±1.68) | 0.431 | ||||
6 | 11 (-7.00±1.26) | 14 (-8.43±1.70) | 0.088 | ||||
12 | 12 (-7.42±1.31) | 9 (-8.56±1.24) | 0.154 | ||||
Last | 9 (-6.78±1.39) | 14 (-8.14±1.51) | 0.012 | ||||
Neck pain score, NRS | 1 | 16 (-0.88±1.50) | 0.751 | 17 (-0.82±4.33) | 0.006 | 0.021 | 0.800 |
3 | 15 (-1.33±1.76) | 17 (-2.59±3.66) | 0.185 | ||||
6 | 11 (-1.00±1.34) | 14 (-2.64±3.50) | 0.103 | ||||
12 | 12 (-1.25±2.42) | 9 (-4.33±4.30) | 0.026 | ||||
Last | 9 (-1.67±2.35) | 14 (-4.36±3.71) | 0.038 | ||||
NDI score | Last | 16 (-19.69±13.68) | 20 (-18.15±20.05) | 0.794 |
Values are presented as number (mean±standard deviation).
Corresponding time value - preoperative value, mean (least squares means) : mean estimated by linear mixed model,
F/U : follow up, ACDF : anterior cervical discectomy and fusion, PCF : posterior cervical fusion, NRS : Numeric rating scale, NDI : Neck disability index