The importance of implant surface characteristics in the replacement of failed implants.

PURPOSE The purpose of the study was to compare the failure rates of implants with either a machined surface or a TiUnite surface used to replace failing implants. MATERIALS AND METHODS The files of 578 patients, ie, of all patients who were treated at the Department of Periodontology of the University Hospital in Leuven by means of oral implants during 3 recent consecutive years, were analyzed. The implants included in the study had an observation time ranging from 9 to 49 months. All patients had been provided with Brånemark System implants. Only 2 types of implant surfaces were used: machined and TiUnite. Data collection and analysis focused on the replacement implants, ie, implants placed at sites where the original implants had failed. Data were statistically analyzed by means of Statistica for Windows Software version 5.1; a Fisher exact P test was used. The level of significance was set at P = .05. RESULTS A total of 41 patients experienced the nonintegration of 58 implants. Of those, 29 implants with a machined surface were replaced by implants with the same surface. Six of the replacement implants failed. Nineteen machined-surface implants were replaced by TiUnite surface implants; 1 failed. Ten TiUnite-surface implants were replaced by implants with the same surface; none failed. The difference in failure rate between machined-surface replacement implants and TiUnite replacement implants was statistically significant (P = .05). DISCUSSION In addition to the usual patient-related compromising factors, replacement of a failing implant involves the challenge of achieving osseointegration in a nonpristine bone site. In the present study, implants with TiUnite surfaces were associated with fewer failures than machined-surface implants under the same conditions. CONCLUSION An improved implant surface such as TiUnite may offer a better prognosis when a failed implant has to be replaced at the same site.