Adjacent Segment Pathology Requiring Reoperation After Anterior Cervical Arthrodesis

STUDY DESIGN: A retrospective study. OBJECTIVE: The purpose of this study was to determine, using survivorship analysis, the rate of adjacent segment pathology (ASP) development and to identify the risk factors for reoperation. SUMMARY OF BACKGROUND DATA: The study of Hilibrand defined "adjacent segment disease" as symptomatic radiculopathy or myelopathy due to an adjacent segment documented on 2 consecutive office visits. In addition to being somewhat subjective, their criterion is not as practical as identifying the rate of adjacent pathology by the need for reoperation. METHODS: This was a retrospective analysis of 1038 consecutive patients who underwent primary anterior cervical spine arthrodesis for radiculopathy and/or myelopathy by 1 surgeon. Annual incidence and prevalence of ASP requiring surgery were calculated and survivorship was determined. We used the Cox regression for risk factor analysis. RESULTS: Secondary surgery on adjacent segments occurred at a relatively constant rate of 2.4% per year (95% confidence interval, 1.9-3.0). The Kaplan-Meier analysis predicted that 22.2% of patients would need reoperation at adjacent segments by 10 years postoperatively. Factors increasing the risk were smoking, female sex, and the number of arthrodesis segments. One or 2-segment arthrodesis had an 1.8 times greater risk than arthrodesis involving 3 or more segments. Age, neurological diagnosis, diabetes, Klippel-Feil syndrome, and noncontiguous segmental-type ossification of posterior longitudinal ligament were not significant risks. CONCLUSION: Patients treated with 1- or 2-segment anterior cervical arthrodesis are more likely to develop ASP than those treated with arthrodesis involving 3 or more segments. Smokers and women had a higher ASP reoperation rate. Our series, the largest in the literature, predicts that 22.2% of patients will require reoperation for ASP within 10 years, substantially higher than the Hilibrand study. LEVEL OF EVIDENCE: 3.