Transition Risks of Complication and Failure for Dental Implants

Objectives: The purpose of this study was to simultaneously predict the risks of transition from complication free to the state of complication or to implant failure based on a set of important predictors.

Material and Methods: A retrospective cohort study design was conducted. The cohort consisted of subjects having at least one Bicon dental implant placed. Among the potential predictors including smoking status, timing of placement, implant staging, age, and gender, only the first three were found significantly associated (p < 0.05) with at least one of the transition risks based on a proportional transition intensity multi-state model (Andersen and Keiding, 2002). The final model for analysis only included these three covariates. The transition probabilities were estimated based on the Breslow-type estimator and their corresponding 95% confidence intervals were constructed based on the bootstrap methods.

Results: The cohort composed of 677 patients with 2349 implants placed. For a one stage, delayed implant placement of a non-smoking patient, the implant would remain complication free with a probability of 0.80 (95% CI: [0.73, 0.86]) by 12 months and 0.56 (95% CI: [0.39, 0.69]) by 60 months. An implant will transit from complication free to the state of complication with a probability of 0.10 ([0.06, 0.16]) by 12 months and 0.19 ([0.09, 0.31]) by 60 months. The probability for a healthy implant to failure without complication was 0.10 ([0.06, 0.14]) by 12 months and 0.26 ([0.12, 0.41]) by 60 months.

Conclusions: Through a multi-state model with information on the smoking status of the patient, timing of the implant placement, and the staging of the implant, we assessed the risks of complication and failure over time. Such predicted risks could potentially assist patient and clinician in decision making on dental implant treatment and management.

Supported by NIH – NIDCR grant R03 DE016919.