Background: Anterior cervical discectomy and fusion (ACDF) using stand-alone cages is an effective method of treating degenerative disease. However, stand-alone cages are reported to have a relatively high incidence of implant subsidence with secondary kyphotic deformity particularly after multilevel ACDF. The purpose of our article was to investigate clinical and radiological outcomes after ACDF using stand-alone cages, at two contiguous levels, with a particular focus on changes in regional alignment and the correlation between alignment of the operated cervical levels and the entire cervical spine.
Methods: Twenty-seven patients with 54 levels and a mean age of 50.8 years were enrolled between January 2005 and August 2006. They underwent ACDF using polyetheretherketone cages packed with demineralized bone matrix without plate fixation at two contiguous levels. Mean follow-up period was 25.5 months (range, 13-60). Clinical outcome was evaluated using two Visual Analog Scales and the Neck Disability Index (NDI). We assessed fusion, regional alignment (RA) of the operated levels and cervical global alignment (GA) preoperatively in the immediate 1-week postoperative period and at the final follow-up. An interspinous distance ≥2 mm was used as an indicator of pseudoarthrosis at each level.
Findings: All patients showed improvements in clinical outcome, with 96% of patients showing mild NDI scores (<14). Radiological solid fusion was obtained at 48 of 54 levels (88.9%) and in 21 of 27 patients (77.8%). Lower cervical levels were significantly more vulnerable to pseudoarthrosis (100%). Fusion rate had no significant correlation with outcome (p > 0.05). RA of the operated levels was improved at the final follow-up compared with preoperatively in 76% of patients, although it had decreased compared with the immediate postoperative period due to subsidence in 84% of patients. In total, 80.8% of patients showed improvements in GA. Furthermore, improvements in RA showed a significant positive correlation with those in GA (p = 0.001), although improvement in RA and GA did not correlate significantly with clinical outcome (p > 0.05).
Conclusions: Though some degree of subsidence occurred in most cases, RA had improved at the last follow-up compared with preoperatively, which contributed to the significant improvement in GA. However, improvement of RA and GA was not correlated with outcomes.