A nomogram prediction of peri-implantitis in treated severe periodontitis patients: A 1-5-year prospective cohort study.

Background: No nomogram of peri-implantitis was reported before which is valuable for risk-estimating, clinical decision-making, and better-patients-communicating. Purpose: To identify the risk indicators and develop a nomogram prediction model of peri-implantitisin treated severe periodontitis patients. Materials and Methods: A prospective study was conducted on 100 patients with 214 implants. Periodontal and peri-implant parameters were evaluated at implant surgery procedure (T1), and at follow-up (T2). Risk factors were analyzed by logistic regression analyses with generalized estimating equations. Nomogram was developed and the discriminatory ability of the model was analyzed. Results: The incidence of peri-implantitis at patient-level and implant level were 16% and 11.2% respectively, with no implant lost. The variables associated with peri-implantitis were the PDT1 >= 6 mm (%) > 10%, the implant position, length, and diameter after adjusting for covariates. A nomogram prediction model of peri-implantitis were developed with factors of PDT1 >= 6 mm (%) > 10% and implant placed in posterior. The area under the ROC curves of stepwise model was 0.794. Conclusions: The residual pockets and implants position were identified as predictors for the peri-implantitis. The nomogram can be used to estimate the risk of peri-implantitis in treated severe periodontitis patients.