A comparison of simulator-tested and -retrieved cervical disc prostheses. Invited submission from the Joint Section Meeting on Disorders of the Spine and Peripheral Nerves, March 2004.

OBJECT: Total joint arthroplasties most commonly fail because the implant becomes worn and a host inflammatory response subsequently develops. Both the material response to the biological environment and the host response to the device must be thoroughly evaluated to establish the efficacy of cervical arthroplasty. Analyses of devices explanted in humans allow evaluation of both responses. Hypothetical wear rates can be determined by comparing in vivo wear with simulator-derived wear. The purpose of this study was to perform explant analyses involving the Bryan and Prestige discs and compare these results with those obtained using spine simulators.
METHODS: Of the approximately 5500 patients treated with the Bryan disc, 11 have undergone explantation of the device. Six of these devices were analyzed for dimensional and chemical changes. Three of the approximately 300 implanted Prestige discs were retrieved, and two were examined microscopically. Histological specimens were assessed for wear particles and host inflammatory response. Additionally, the extent of simulator-produced wear was compared with that demonstrated in the retrieved specimens.
CONCLUSIONS: The simulator-generated results predict adequate wear-related characteristics for both the Bryan and Prestige prostheses for a minimum of 40 years. Comparison of data with those of the retrieved specimens indicates that the wear was more minimal than predicted in simulators by five- to 10-fold. In no instance did the revisions result from failure of the device due to a reaction to wear debris, fracture, polymer oxidation, or metal corrosion. The inflammatory response seen in the periprosthetic tissues was minimal and not characteristic of inflammatory responses in failed diarthrodial joint arthroplasties.