Comparison of adverse events between cervical disc arthroplasty and anterior cervical discectomy and fusion: a 10-year follow-up

BACKGROUND CONTEXT: Cervical disc arthroplasty (CDA) has been advocated as an alternative to anterior cervical discectomy and fusion (ACDF) with the added potential to reduce the risk of adjacent level disc degeneration and segmental instability. However, the long-term adverse events of arthroplasty have yet to be fully reported. PURPOSE: To investigate the 10-year follow-up adverse events rates between CDA and ACDF. STUDY DESIGN/SETTING: The study was a randomized, prospective, multicenter Investigational Device Exemption (IDE) trial and its continued follow-up as a postapproval study (PAS). Single level surgeries were performed for cervical disc pathologies between May 2002 and October 2004. PATIENT SAMPLE: n=463 patients. OUTCOME MEASURES: Adverse events comparison of CDA and ACDF from self-reported and physiologic measures. METHODS: At each evaluation time point, subjects were queried for adverse events since their last visit; and all adverse events were documented, regardless of whether or not they appeared related to the surgery or device. Adverse events were recorded, categorized, and assessed for severity and relationship to the study device and/or surgical procedure. The 10-year cumulative rates for each type of adverse events were summarized using a life-table method for the time-to-event analysis. A log-rank test was used to compare the two treatment groups. RESULTS: A total of 242 patients received CDA and 221 patients received ACDF. At 10-year follow-up, 54% (130/242) of CDA patients and 47% (104/221) of the ACDF subjects were evaluated. At up to 10-year follow-up, 231 patients in the CDA group (cumulative rate 98.4%) and 199 patients in the ACDF group (cumulative rate 98.7%) had at least one adverse event. Overall, the difference in the cumulative rate of all adverse events over 10 years was not statistically different (p=0.166). The cumulative rates of the following adverse events were not different between the two groups for cancer, cardiovascular, death, dysphonia/dysphagia, gastrointestinal, infection, urogenital, respiratory, implant displacement/loosening, implant malposition, neck and arm pain, neurological, other pain, spinal events, and intraoperative vascular injury. However, there were more adverse events in the CDA group resulting from trauma (p=.012) and more spinal events at the index level (p=.006). The ACDF group had significantly more nonunion events (p=.019), and nonunion outcome pending (p=.034), adjacent level spinal events (p=.033), and events that fell into the “other” category (p=.015). CONCLUSIONS: The cumulative rates of patients who had any adverse events were not different between the artificial cervical disc and the anterior cervical arthrodesis groups. In addition, the cumulative rates were not different between the two groups for the majority of categories as well.