Survival of turned and roughened dental implants in irradiated head and neck cancer patients: A retrospective analysis

Dental implants placed into previously radiated fields in survivors of head and neck cancer can demonstrate survival rates that are less than optimal. Understanding this behavior may assist with treatment planning in accordance with expected survival rates in these patients. The purpose of this study was to compare the survival of implants with turned and roughened surfaces placed in patients who had previously received head and neck radiation, and to identify factors associated with implant failure. The records of 48 patients who had prior head and neck radiation and had 271 dental implants placed between May 1987 and July 2008 were examined. All of the implants were placed in a previously irradiated field dosed to at least 50 Gy. Implant survival was estimated using the Kaplan-Meier method and univariate Cox models with robust standard errors were fitted to evaluate the association between patient/implant factors and implant failure. There were 62 implants placed in the maxilla (32 roughened, 30 turned) and 209 implants (107 roughened, 102 turned) placed in the mandible. The 5-year implant survival rate for implants placed in the maxilla was 72.6 for turned implants and 87.5 for roughened implants. For implants placed in the mandible, the 5-year survival rate was 91.7 for turned implants and 100 for roughened implants. Among implants with a turned surface, implants were more likely to fail if they were placed in the maxilla (P=.008) and if the diameter was ≤3.75 mm (P=.002). Implants with turned surfaces were 2.9 times more likely to fail compared to the roughened dental implants, although the difference did not reach statistical significance. For implants with turned surfaces, there was a tendency for implants in the maxilla to fail more frequently compared to the mandible. Implants with turned surfaces had a higher likelihood of failure in the posterior region than in the anterior region. For implants with roughened surface there was no significant association between implant survival and location of the implant, type of bone, or length or diameter of the implant.