Investigating the Factors Affecting Plaque Formation after Peri-Implantitis and Incorporating Crucial Factors to...

This study aims to explore the factors impacting implant plaque re-formation after implant polishing surgery for peri-implantitis and to establish a predictive model using crucial factors, thereby providing an evidence-based reference for managing this condition.

This retrospective study analyzed clinical data from 203 patients who underwent implant polishing and shaping procedures in Suzhou Stomatological Hospital between November 2018 and October 2023. Study subjects were divided into a training set (n = 142) and a validation set (n = 61) in a 7:3 ratio. Univariate and multivariate logistic regression analyses were used to assess the risk factors associated with biofilm formation after implant polishing and shaping surgery. Incorporating significantly linked factors, a risk prediction model was developed. Furthermore, the predictive model was evaluated in the training and validation sets using the Hosmer-Lemeshow (H-L) goodness-of-fit test, Receiver Operating Characteristic (ROC) curve, calibration curve, and decision curve analysis (DCA) to determine its discriminatory capability, goodness-of-fit, and predictive utility.

Multivariate logistic regression analysis revealed that calculus [Odds Ratio (OR) = 3.071, 95% Confidence Interval (CI): 1.104-8.541, p = 0.032], difficult-to-clean implant location (OR = 5.807, 95% CI: 1.895-17.798, p = 0.002), external connection implant abutment (OR = 4.378, 95% CI: 1.440-13.308, p = 0.009), and implant diameter (OR = 4.511, 95% CI: 2.141-9.504, p < 0.001) were significant factors affecting biofilm formation after implant polishing and shaping surgery. A regression equation (predictive model) was constructed, incorporating the four crucial risk factors and regression coefficients. ROC curve analysis demonstrated that the area under curve (AUC) of the predictive model was 0.9143 (95% CI: 0.8221-0.9782) in the training set and 0.8095 (95% CI: 0.7342-0.9051) in the validation set. Furthermore, the Hosmer-Lemeshow test indicated a good fit of the established model, with no statistically significant difference between predicted and observed values in both the training set (p = 0.399) and the validation set (p = 0.317). Additionally, DCA demonstrated that the predictive model provides a significant net benefit.

The predictive model developed using the key risk factors contributing to plaque formation after implant polishing exhibits strong predictive capability, which provides an evidence-based reference in preventing and managing postoperative plaque formation.