Collar Fixation Is Not Mandatory After Cervical Laminoplasty

STUDY DESIGN: Prospective randomized controlled study. OBJECTIVES: To determine the effect of collar-aided fixation on outcomes after laminoplasty for cervical compressive myelopathy. SUMMARY OF BACKGROUND DATA: Patients are often placed in a Philadelphia collar for weeks after cervical laminoplasty. However, the benefit of postoperative collar use remains controversial. We hypothesized that treatment outcomes in postoperative pain in patients without collars would not be inferior to the outcomes in patients using Philadelphia collars for 2 weeks. METHODS: This trial analyzed 74 patients (52 males, 22 females, mean age 72.7 years) with cervical compressive myelopathy who had undergone double-door laminoplasty. Before surgery, we randomly assigned patients to the collar-fixation (CF) group (postoperative Philadelphia collar for 2 weeks) or the no-collar (NC) group. The primary outcome was the visual analog scale (VAS) for cervical pain up to 1 year after surgery. Secondary outcomes were Japanese Orthopedics Association score, Short Form 36 score (SF-36), cervical range of motion (ROM), lordotic angle, and complications. RESULTS: VAS scores up to 1 year after operation were similar with or without collar fixation (P = 0.487, two-way ANOVA). JOA scores improved significantly in both groups (P = 0.002 in CF, P < 0.001 in NC). There was no statistically significant difference between groups with regard to the improvement of JOA scores (54.9% in CF, 47.0% in NC, P = 0.80). The improvement in SF-36 was similar in both groups. Loss of ROM and lordotic angle of the cervical spine did not differ between groups (P = 0.61 in CF, P = 0.82 in NC). The incidence of complications was similar in both groups. CONCLUSION: The VAS scores of cervical pain with the postoperative treatment without collar fixation were not inferior to those when using Philadelphia collars for 2 weeks. Moreover, other outcomes such as JOA scores, SF-36, ROM, lordotic angle, and complications were similar in both groups. LEVEL OF EVIDENCE: 2.