We retrospectively reviewed data from patients treated for ACDF using the Zero-P in our clinic between January 2013 and December 2016. A total of 53 patients were enrolled in this study. Of the 53 patients, 17 were treated with only Zero-P, i.e., without posterior fixation for traumatic disc injury (group A), and the remaining 36 patients were treated for degenerative disc herniation (group B). Surgical indicated levels were a single segment or skipped two segments (not including contiguous two segments, such as C3/4/5), determined by evaluating the preoperative radiologic pathology and clinical symptoms. Patients who met any of the following criteria were excluded in group A : 1) vertebral body fracture, 2) facet dislocation, and 3) cases requiring posterior decompression and instrumentation, such as severe canal stenosis and kyphotic deformity. In addition, the following criteria were also excluded in group A and B as well : 1) prior cervical spine surgery, 2) autoimmune or metabolic bone disease, and 3) postoperative surgical site infection.

In group A, the presence of a paravertebral ligament injury, prevertebral hemorrhage, facet perch and the score from the Sub-axial Cervical Spine Injury Classification System (SLICS) were further evaluated through medical charts, operation records, and preoperative magnetic resonance imaging (Fig. 1).

For ACDF using Zero-P, the patient was placed in the supine position under general anesthesia. Fluoroscopy was used to confirm the target level. A horizontal incision was made along the skin crease in the neck that correlated to the target cervical disc. After removal of the intervertebral disc with a careful endplate preparation, generous decompression of the bilateral neuroforamen was performed. A Kerrison punch and high-speed electric drill were used to decompress the nerve roots by removing the osteophyte overgrowth on the uncovertebral joint and posterior lips of the vertebral body. Extensive saline irrigation was performed during drilling of the osteophytes and milling of the endplates. After sufficient decompression, we applied a Zero-P convex implant and 4 Zero-P screws under intraoperative fluoroscopy. After the release of a Casper distractor, a manual pullout test was conducted to confirm the stability of the segments. A closed system drainage catheter was placed. After surgery, all patients were instructed to wear a Philadelphia neck brace for 4 weeks.

Radiological images were assessed pre-operation and at follow-ups that occurred approximately 3, 6, and 12 months postoperatively. The cervical curvature angle, fusion status, and subsidence rate was assessed using cervical X-ray and cervical computed tomography (CT). All radiographic reports were interpreted and measured using PACS system software (INFINITT, Seoul, Korea).

The cervical Cobb angle was measured as the acute angle constructed by the lines running along each inferior vertebral endplate of the C2 and C7 vertebral bodies on lateral X-ray. The operative segmental angle was measured as the acute angle constructed by the lines running along the superior and inferior vertebral endplates of the operated vertebral body. The difference between the pre-operative and 12-month postoperative follow-up angles for each parameter was designated as the delta (Δ) value (Fig. 2).

The evaluation of fusion of the operated cervical vertebral body was determined using dynamic cervical X-ray 12 months postoperatively. In cases in which it was difficult to determine the fusion status accurately by plain radiograph, the trabecular bone bridge was checked using cervical CT (Fig. 3). The protocol for bone fusion was as follows : 1) magnify image >150%; 2) assess for adequate flexion-extension effort by confirming that there is >4 mm of inter-spinous process motion at a non-arthrodesed level; and 3) ensure there is <1 mm of inter-spinous process motion at the fused level [17].

Cage subsidence was defined as a decrease in interbody height >3 mm on a plain radiograph 12 months postoperatively. The difference in subsidence was based on the length between the center of the upper and lower endplates in a lateral X-ray.

The clinical outcomes were evaluated pre- and postoperatively and at the last follow-up using the Japanese Orthopedic Association (JOA) score. At the last follow-up, patients were assessed according to Odom’s criteria, which ranging from ratings of excellent to poor [12].

Patient characteristics and clinical outcomes were assessed using descriptive analyses, with the mean (standard deviation) and median (interquartile range) presented for quantitative variables and the frequency (percent) presented for qualitative variables. The comparison of characteristics and clinical outcomes between groups A and B were analyzed using twosample t-tests or Mann-Whitney U tests for quantitative variables and chi-square tests or Fisher’s exact test for qualitative variables. p-values are presented for all statistically significant variables. All statistical analyses were performed using the SPSS software package for Windows version 19.0 (IBM, Chicago, IL, USA). All tests were 2-sided, and p-values <0.05 were considered to indicate statistical significance.

The fusion rate was 95.0% (19/20) in group A and 95.1% (39/41) in group B. The mean Cobb and operative segmental angles increased by 3.45 mm and 2.94 mm, respectively, over the preoperative value in group A. Similarly, the Cobb and operative segmental angles increased by 2.46 mm and 2.88 mm, respectively, in group B. Cage subsidence occurred in 7 levels (7/20, 35.0%) in group A and 12 levels (12/41, 29.3%) in group B. None of the measured values were significantly different between the two groups (all p>0.05, Table 3).

Both groups exhibited an improvement in clinical outcomes as demonstrated by the increase from the mean preoperative JOA score to the postoperative JOA scores (JOA; group A, 1.59 ±1.58; group B, 1.44±1.05). There were no statistically significant differences with regard to the JOA values between the two groups (p=0.696). According to Odom’s criteria, all outcomes had a grade higher than “good” in both groups. There vwere no significant differences in the proportions of “good” outcomes between the two groups (p=0.667, Table 4).

We performed a comparative analysis of male subjects in both groups, because group A was mostly consisted of male patients (16/17, 94.1%) and the number was similar to male patients in group B. In terms of radiologic and clinical outcomes, there were no significant differences between male subjects of group A and B. Detailed data for comparisons of both groups are presented in Table 5.

Three patients (two in group A and one in group B) suffered from transient dysphagia, but recovered within 3 weeks postoperation. In group B, one patient reported prolonged vocal cord paralysis. No complications requiring re-operation, such as surgical site infection, instrumental failure, or postoperative hematoma, occurred.