Object. The authors report the preservation of motion at surgically treated and adjacent spinal segments after placing an artificial cervical joint (ACJ) and they describe the influence of interbody fusion on changes in angulation occurring in the sagittal plane at adjacent levels in the treatment of cervical spondylosis. Methods. The authors conducted a prospective nonrandomized study of patients in whom an ACJ was placed or autologous bone graft interbody fusion was performed. Angular measurements at levels adjacent to that surgically treated were calculated using plain flexion—extension radiographs obtained at 6-month intervals. Analyses of qualitative data, such as increase or decrease in adjacent-level motion, and the degree of disc degeneration were performed. Quantitative data were also analyzed. In the fusion group a significant increase in adjacent-level movement was demonstrated at the 12-month follow-up visit compared with the group of patients in whom ACJs were placed (p < 0.001). The increase in movement occurred predominantly at intervertebral discs that were preoperatively regarded as normal (p < 0.02). An overall reduction in adjacent-level movement was observed in patients who underwent joint replacement, although this was compensated for by the movement provided by the ACJ itself. Conclusions. Fusion results in increased motion at adjacent levels. The increase in adjacent-level motion derives from those discs that appear radiologically normal prior to surgery. It remains unknown whether ACJs have a protective influence on adjacent intervertebral discs.

There is limited evidence supporting the use of a cervical interbody fusion device in place of autologous bone. There is a need to standardize the testing of implants with good quality laboratory work preceding clinical use. Certain devices including cages, some forms of hydroxyapatite, and bone morphogenic proteins merit further study.

A prospective randomized study was undertaken to evaluate the radiological appearance and clinical effectiveness of two porous tantalum (Hedrocel) implants in achieving a stable cervical interbody fusion. A prerandomization protocol was used to allocate patients to the three arms of the study: a ring implant containing autologous cancellous bone graft, a solid block implant or autologous tricortical iliac crest bone graft. Patients were followed for 2 years with plain radiological studies, SF-36, and Neck Disability Index questionnaires and neurological assessment. Early in the study the postoperative radiographs of four patients receiving Hedrocel implants showed inferior end-plate lucency raising concerns about delayed or non-fusion. Recruitment to the study was halted by the investigators to allow longer-term follow-up of the implanted patients when only 24 patients had been recruited to the study. Although fusion was subsequently noted in all patients at 12 months there was no further enrolment to the study. At 2 years the radiological and clinical outcomes of the three groups appeared comparable, but the study numbers were too small for any statistical analysis. This study highlights the difficulties that can arise when clinical caution takes precedence over objective measures of clinical progress during a study. In the absence of an independent safety monitoring committee, the investigators were under an ethical obligation to suspend recruitment to this study, until it was clear that the radiological features were not associated with poor clinical outcomes. The use of safety monitoring committees and the clarification of stopping criteria in relation to outcome measures should be considered in open randomized trials of spinal surgical techniques and implants.