Exploratory analysis of predictive factors for radiographic repair of endo-periodontal lesions: a pilot study

Abstract Objective: This study aimed to evaluate factors associated with the radiographic repair of EPLs, with a particular focus on clinical, periodontal, and systemic variables, using a decision tree analysis to identify the most influential factors. Material and method: This retrospective study analyzed records of patients treated for EPLs at UNIFAE (2022-2023), who received combined endodontic-periodontal therapy and were reassessed at least six months post-treatment. Clinical, demographic, and radiographic data were collected, including probing depth (PD), bleeding on probing (BOP), diabetes status, age, and follow-up duration. Patients were classified into two categories based on radiographic and clinical criteria: complete repair or partial repair. Statistical analyses included chi-square and t-tests, complemented by a decision tree model (CART algorithm) to identify the primary predictors of radiographic repair. Result: The study initially included 16 patients (18 teeth), but follow-up was completed with only seven patients (seven teeth), with longer follow-up linked to complete radiographic repair. The decision tree analysis confirmed follow-up time as the strongest predictor, followed by diabetes and PD. Patients with diabetes exhibited a higher likelihood of partial repair. Although bivariate analyses did not demonstrate statistical differences, the exploratory model highlighted the potential hierarchy of these variables. Conclusion: These findings suggest the importance of prolonged follow-up and systemic health, especially diabetes, in the prognosis of EPL treatment, emphasizing the need for individualized monitoring and integrated endodontic-periodontal care. However, these findings should be interpreted with caution due to the limited sample size. Further large-scale, prospective studies are needed to confirm these predictors.