Background: We studied the incidence, prevalence, and radiographic progression of symptomatic adjacent-segment disease, which we defined as the development of new radiculopathy or myelopathy referable to a motion segment adjacent to the site of a previous anterior arthrodesis of the cervical spine.Methods: A consecutive series of 374 patients who had a total of 409 anterior cervical arthrodeses for the treatment of cervical spondylosis with radiculopathy or myelopathy, or both, were followed for a maximum of twenty-one years after the operation. The annual incidence of symptomatic adjacent-segment disease was defined as the percentage of patients who had been disease-free at the start of a given year of follow-up in whom new disease developed during that year. The prevalence was defined as the percentage of all patients in whom symptomatic adjacent-segment disease developed within a given period of follow-up. The natural history of the disease was predicted with use of a Kaplan-Meier survivorship analysis. The hypothesis that new disease at an adjacent level is more likely to develop following a multilevel arthrodesis than it is following a single-level arthrodesis was tested with logistic regression.Results: Symptomatic adjacent-segment disease occurred at a relatively constant incidence of 2.9 percent per year (range, 0.0 to 4.8 percent per year) during the ten years after the operation. Survivorship analysis predicted that 25.6 percent of the patients (95 percent confidence interval, 20 to 32 percent) who had an anterior cervical arthrodesis would have new disease at an adjacent level within ten years after the operation. There were highly significant differences among the motion segments with regard to the likelihood of symptomatic adjacent-segment disease (p < 0.0001); the greatest risk was at the interspaces between the fifth and sixth and between the sixth and seventh cervical vertebrae. Contrary to our hypothesis, we found that the risk of new disease at an adjacent level was significantly lower following a multilevel arthrodesis than it was following a single-level arthrodesis (p < 0.001). More than two-thirds of all patients in whom the new disease developed had failure of nonoperative management and needed additional operative procedures.Conclusions: Symptomatic adjacent-segment disease may affect more than one-fourth of all patients within ten years after an anterior cervical arthrodesis. A single-level arthrodesis involving the fifth or sixth cervical vertebra and preexisting radiographic evidence of degeneration at adjacent levels appear to be the greatest risk factors for new disease. Therefore, we believe that all degenerated segments causing radiculopathy or myelopathy should be included in an anterior cervical arthrodesis. Although our findings suggest that symptomatic adjacent-segment disease is the result of progressive spondylosis, patients should be informed of the substantial possibility that new disease will develop at an adjacent level over the long term.