Study Design. A retrospective study. Objective. To assess the long-term results of zero-profile spacer for 3-level anterior cervical discectomy and fusion (ACDF). Summary of Background Data. Although widely used, there are still controversies about the long-term results of zero-profile spacer, especially in multilevel cases. Methods. Cases received 3-level ACDF for cervical spondylotic myelopathy (CSM) using either zero-profile spacer (n = 27) (ZP Group), or plate and cages (n = 34) (PC Group), and with 5-year follow-up were reviewed. Neurological function and life quality were assessed by modified Japanese Orthopaedic Association (mJOA) score, Neck Disability Index (NDI), and Short-Form 36 (SF-36) score. Disc height, cervical lordosis, fusion rate, and surgical complications were observed. Results. Neurological recovery and life quality improvement were similar in both groups. Disc height and cervical lordosis (C2-7 Cobb angle) were well restored after operations, but lost in both groups during follow-up. Loss of correction (LOC) in disc height was larger in ZP Group (11.38% vs 5.71%, P < 0.05) at 5-year follow-up. LOC of cervical lordosis in ZP group constantly grew from 11.28% to 48.13% during 5-year follow-up, significantly higher than that in the PC group (from 7.43% to 14.01%) (P < 0.05). The rate of postoperative dysphagia was no statistical difference between the two groups, and symptoms were all disappeared within 1 year. There were 10 levels of adjacent segment degeneration (1 in ZP Group, and 10 in PC Group, P = 0.02). Cage subsidence (11 of 81 levels, 13.58%) and screw migration (2 of 81 levels, 2.47%) were only observed in the ZP Group. The migrated screws in one case were surgically removed. Fusion was achieved in all cases. Conclusions. In long-term follow-up of 3-level ACDF for CSM, zero-profile spacer has the similar clinical results, but loss of correction of disc height and cervical alignment were significantly higher, compared with anterior plate and cages. Level of Evidence: 3

No abstract