Optimal Time Period of Wearing Protective Collar After Anterior Cervical Discectomy and Fusion


 Background: There is still no consensus on the time period of wearing collar after anterior cervical discectomy and fusion (ACDF). We aim to investigate the optimal time period of wearing protective collar.Methods: We retrospectively reviewed patients with cervical spondylosis who underwent one to two segment ACDF during January 2016 and December 2017, and included 97 patients who meet inclusion and exclusion criterion. Patients were divided into three groups according to the actual time period of wearing collar after ACDF including 1-4 week group, 5-8 week group, and 9-12 week group. We analyzed Japanese Orthopedic Association (JOA) score, Axial Symptom (AS) score and Neck Disability Index (NDI) before surgery and at post-operative 3 months to investigate the optimal time period of wearing collar.Results: JOA score: All three groups have a better post-operative JOA score compared with that before surgery (paired t test, p<0.05). There is no significant difference among the three groups with respect to post-operative JOA (ANOVA, p>0.05).AS score: The post-operative AS scores of 1-4 week group and 5-8 week group were significantly better than that before surgery (paired t test, p>0.05). While the post-operative AS score of 9-12 week group was significantly worse than preoperative AS score (paired t test, p<0.05).NDI: All three groups have a better post-operative NDI compared with that before surgery (McNemar test, p<0.05). Of note, in 5-8 week group, the percentage of no deficit increased by 45%, and the percentage of mild deficit decreased by 45% accordingly. That percentage is 26% and 31% in 1-4 week group and 9-12 week group, respectively. There was significant difference among these three groups (Fisher's exact probability test, p<0.05)Conclusions: For cervical spondylosis patients who underwent 1-2 segment ACDF, the optimal time period of wearing protective is 5-8 weeks. This time period results in comparable neurological outcome, least axial symptom risk, and highest chance of no deficit on neck function.

AS score: The post-operative AS scores of 1-4 week group and 5-8 week group were signi cantly better than that before surgery (paired t test, p>0.05). While the post-operative AS score of 9-12 week group was signi cantly worse than preoperative AS score (paired t test, p<0.05).
NDI: All three groups have a better post-operative NDI compared with that before surgery (McNemar test, p<0.05). Of note, in 5-8 week group, the percentage of no de cit increased by 45%, and the percentage of mild de cit decreased by 45% accordingly. That percentage is 26% and 31% in 1-4 week group and 9-12 week group, respectively. There was signi cant difference among these three groups (Fisher's exact probability test, p<0.05) Conclusions: For cervical spondylosis patients who underwent 1-2 segment ACDF, the optimal time period of wearing protective is 5-8 weeks. This time period results in comparable neurological outcome, least axial symptom risk, and highest chance of no de cit on neck function.

Background
Cervical spondylosis is a common degenerative disease of the cervical spine. [1] Degeneration of the cervical intervertebral disc and its secondary pathological changes cause stimulation or compression on adjacent nerve roots, spinal cord, vertebral artery, or cervical sympathetic nerves, resulting in various symptoms such as neck pain, numbness and weakness of limbs. With the changes in people's life and work style, especially popularization of computer and smart phone, more and more people experience long-term sub-healthy cervical spine, which eventually leads to cervical spondylosis. Min et al. reported nearly 150 million people in China suffering from cervical spondylosis, including 82% of people over 60 years old, 71% of people aged 50-60 years old and 59.1% of young adults aged 30-40 years old. [2] For the treatment of cervical spondylosis, conservative treatments such as lifestyle modi cation, pain relief, and physical therapy should be tried rstly. If the symptoms are severe and refractory or myelopathy occurs, surgery is recommended. Since the anterior cervical discectomy and fusion (ACDF) was rst proposed by Cloard and Robinson in the mid-20th century, this procedure has been widely used in clinical practice and has become a common operation for the treatment of cervical degenerative diseases. [3] Wearing a protective collar is usually recommend after ACDF to maintain the stability of the spine, promote intervertebral fusion, and provide patients with a sense of security in daily activities. However, there is still no consensus on the time period of wearing collar. There are different opinions in the literature on the time period of wearing collar after ACDF. Several studies have shown that 1-2 segment ACDF does not require postoperative external cervical support, [4][5][6] and other literature reported that patients who underwent 1-2 segment ACDF needed wearing collar for 4-12 weeks with off-bed ambulation [7][8][9][10][11]. Therefore, it is very important to study the impact of different time period of wearing cervical collar after ACDF on the safety, comfort and prognosis of patients. We conducted this retrospective study to explore the optimal time period of wearing collar for patients received ACDF.

Methods
Patients with cervical spondylosis who were admitted to our center from January 2016 through  Table 1 lists demographic details for the three groups. There were 39 patients in the 1-4 week group, 22 in the 5-8 week group, and 36 in the 9-12 week group. There was no statistical difference in age, gender, BMI, drinking history, smoking history, hypertension, diabetes, coronary heart disease among the three groups of patients (P>0.05), indicating that the three groups of patients are comparable.  JOA score: The results of intra-group analysis showed that the JOA scores of the three groups after operation were signi cantly improved compared with those before operation (paired t-test, p<0.05). Since there was no difference in the preoperative JOA baseline of the three groups (ANOVA analysis, P>0.05), postoperative JOA score of each group can be further compared. There is no statistical difference between the three groups (ANOVA analysis, P> 0.05), indicating that the time period of wearing collar has no signi cant effect on neurological rehabilitation.
AS score: Intra-group analysis showed that the postoperative AS scores of the patients in the 1-4 week group and the 5-8 week group increased compared with those at baseline (paired t test, p<0.05). While the postoperative AS score of 9-12 week group decreased (paired t test , P<0.05), suggesting that wearing collar for 9-12 weeks will increase risk of axial symptoms. Since there was no difference in the baseline AS scores of the three time periods (ANOVA analysis, P>0.05), the postoperative AS scores of each group were further compared. The AS score of 9-12 week group was lower than that of 1-4 week group (ANOVA Analysis, Tukey's test, P<0.05), while there was no statistically signi cant difference between the other groups (ANOVA analysis, Tukey's test, P>0.05). The incidence of axial symptoms of patients in the 1-4 week group, 5-8 week group, and 9-12 week group were 10.3%, 13.6%, and 22.2%, respectively. The incidence of new axial symptoms was 19.4% (7/36) for 9-12 week group. Therefore, the time period for wearing collar should not exceed 8 weeks.
NDI score: Intra-group analysis showed that the postoperative NDI scores of the three groups were signi cantly improved compared with that preoperatively (McNemar test, P<0.05). Since the baseline of the preoperative NDI of the three groups was comparable (chi-square test, P>0.05), inter-group comparison of postoperative NDI scores between groups can be made. In 5-8 week group, patients with no de cit (as per NDI score) increased by 45% post-operatively, and the proportion of patients with mild de cit decreased by 45% accordingly, which is signi cantly better that that of 1-4 week group (26%) and 9-12 week group (31%). (Figure 2)

Discussion
There is no consensus on the time period of wearing protective collar after ACDF. Short wearing time may be criticized for poor spinal stability and low fusion rate. On the contrary, patients who wear it for a long time may experience discomfort or even complications such as axial symptoms, in which cervical muscle atrophy, stiffness and contracture of the ligament and joint capsule tissue leading to neck pain, soreness, and stiffness. [7] Neck pressure may interfere with normal lymphatic and venous return, causing tissue edema after surgery and nally airway obstruction. [7] The current common practice for a surgeon is to dictate a patient to wear collar according to their personal experience, rather than a consensus or evidence-based recommendation. This study retrospectively analyzed the actual time period of wearing collar and the clinical outcomes of 97 patients who underwent one or two segment ACDF procedure and found that the optimal time period for wearing protective collar after ACDF is 5-8 weeks, which can provide the best balance of safety and comfort for the patient.
We analyzed the commonly used JOA, AS, NDI scores according different time period (1-4 week, 5-8 week, and 9-12 week) of wearing protective collar after ACDF. Comparison of JOA scores across the three groups shows that different time period has no statistically signi cant difference as per the rehabilitation of nerve function. Analyzing AS scores across the three group shows that AS score in 9-12 week group is signi cantly reduced compared with 1-4 week or 5-8 week group. The incidence of axial symptoms was 10.3%, 13.6% and 22.2% in 1-4 week, 5-8 week and 9-12 week group, respectively (p < 0.05, ANOVA). Therefore, from the AS score perspective, time period of wearing collar after ACDF should not exceed 8 weeks so as not to increase the risk of axial symptoms. Finally, according to NDI scores among the three groups, the percentage of patients with no de cit increased by 26%, 45% and 31% in 1-4 week, 5-8 week and 9-12 week group, respectively (p < 0.05, Fisher exact test). To sum up, these results show that the optimal time period of wearing protective collar after ACDF is 5-8 weeks. This time period results in comparable neurological outcome (as per JOA score), least axial symptom risk (as per AS score), and highest chance of no de cit on neck function (as per NDI score). wearing a neck brace for 6 weeks after surgery can signi cantly reduce NDI and the level of neck pain. [8] There are also studies that support usage of protective collar for 4-12 weeks after 1-2 level ACDF [9][10][11], which is partially consistent with the optimal 5-8 week wearing time in this study.
Several limitations exist in literature and our study. There is enormous heterogeneity among studies published, which accounts for different conclusions. Our study has a relatively small sample size, and time period partition is some extent arbitrary. Further studies with large sample size, prospective design, and more re ned time periods are warranted.

Conclusions
Our study showed that the optimal time period of wearing protective collar after 1-2 segment ACDF is 5-8 weeks. Availability of data and materials All data generated or analysed during this study are included in this published article and its supplementary information les.

Competing interests
The authors declare that they have no competing interests.   Pre-versus post-operative NDI score in each group. NDI, Neck Disability Index. Pre-, preoperative. Post-, postoperative 3 months.