Risk Factors for Implant Failure and Peri‐Implantitis: A Retrospective Cohort Study
Nathalia Vilela, Bruno C. V. Gurgel, Hélio D. P. da Silva, Claudio M. Pannuti, Poliana M. DuarteABSTRACT
Objectives
This retrospective cohort study evaluated patient‐, implant‐, site‐, procedural‐, biological‐, and prosthesis/mechanics‐related factors associated with early and late implant failure and peri‐implantitis.
Material and Methods
Electronic health record data from a university center were analyzed. Forty patient‐, implant/site‐, biologic‐, and prosthesis‐related variables were evaluated in relation to early failure, late failure, and peri‐implantitis. Bivariate and survival analyses were performed, followed by mixed‐effects regression models appropriate to each outcome. Robustness was assessed via clustered bootstrap resampling, penalized regression, and sensitivity analyses.
Results
A total of 264 patients and 554 implants were assessed. Cumulative implant failure was 6.8% (3.4% early; 3.4% late), and peri‐implantitis incidence was 7.6%. In multivariable analyses, self‐reported penicillin allergy (OR = 7.9; 95% CI: 2.7–23.7) and anterior implant location (OR = 5.4; 95% CI: 2.0–14.0) were associated with higher odds of early implant failure, whereas antihypertensive medication use (OR = 0.26; 95% CI:0.08–0.91) was associated with lower odds ( p < 0.05). Late implant failure was associated with hypertension (HR = 2.1; 95% CI: 1.3–3.3), previously grafted sites (HR = 1.4; 95% CI:1.0–1.9), and bone‐level implants (HR = 10.4; 95% CI: 5.5–19.5). In cluster‐robust sensitivity analysis, the association between bone‐level implants and late implant failure was attenuated but remained significant (HR = 3.38; 95% CI: 1.96–5.84). Peri‐implantitis was associated with periodontitis (HR = 3.6; 95% CI: 1.89–6.97), excess cement (HR = 2.7; 95% CI: 1.25–5.88), and PLCBL (HR = 2.1; 95% CI: 1.09–4.13) ( p < 0.05). Robustness analyses generally supported the direction of the identified associations, although some effect estimates were attenuated.
Conclusions
Implant failure and peri‐implantitis are associated with distinct, potentially modifiable patient‐ and treatment‐related factors. Recognition of these factors may help identify patients and implants at increased risk of adverse outcomes.