Purpose
The basic aim of the present study was to analyze the results of implant failures in two different implant populations, and how these results may vary depending on how the data are compiled and analyzed.
Materials and Methods
Two groups of consecutively treated patients were included who had been provided with either one single implant in a partially edentulous upper jaw (1881 patients) or four to eight implants in an edentulous upper jaw (2031 patients/12 454 implants). The risk of implant failure in the two groups separately and in combination was statistically compared by using uni‐ and multivariable analyses.
Results
The two groups showed significant differences in inclusion, surgical treatment protocols, and the risk of implant failures (P < .05). Overall, 25‐year patient‐level cumulative survival rates (CSRs) were 75.8% and 96.3% for edentulous and single implant treatment, respectively. “Dental condition” was the variable associated with the greatest risk of implant failure (HR 6.00; edentulous). Only one variable was significantly associated with the risk of implant failure in all tested groups (“time after surgery”; a decreased risk was observed over time), and more variables were statistically associated with implant failures in the edentulous group than in the single implant group.
Conclusions
Edentulous patients present a significantly and substantially higher risk of implant failures than patients provided with a single implant. When patients with different clinical conditions are pooled into the same group, patients with the most common condition in the total group have greatest impact on the result of the total group. Based on the present observations, risk patterns for a certain oral condition are not necessarily comparable with the implant treatment received by other patients, and the external validity may be limited in small, homogeneous groups of patients.