Logo do repositório

Percorrer por autor "Beato, Frederico Marcelo Rodrigues"

A mostrar 1 - 1 de 1
  • Validation of a predictive model of early tooth loss in periodontitis patients
    Publication . Beato, Frederico Marcelo Rodrigues; Mendes, José João; Proença, Luís
    Introduction: Periodontitis is a highly prevalent oral disease characterized by tissue destruction with loss of the dentition’s attachment apparatus and ultimately results in tooth loss, if not treated. Patients with periodontitis and tooth loss see their quality of life greatly impacted. The continuing growth of precision dentistry and prediction modeling approaches to Periodontology aims to excel treatments while minimizing unnecessary costs in what ends up being a more informed and tailored approach, improving the patient's quality of life by providing more effective and predicted treatments. Importantly, such models do not exist for the Portuguese population and this study aims to validate a previously developed prediction model for early tooth loss in periodontitis patients, defined as extractions after diagnosis and before periodontal treatment. Materials and methods: A previously developed model with the following equation: Log[Prob. (Ext.)/(1-Prob. (Ext.))] = -7.850 + 0.589 × TD(Incisive) + 0.661 × CAL (clinical attachment loss) was used. The validation sample involved a retrospective cohort of patients from the Periodontology Department of the Egas Moniz Dental Clinic, and recruited consecutively from June 2018 to March 2020. The validation process was carried out following the Transparent Reporting of a multivariable prediction model for Individual Prognosis Or Diagnosis (TRIPOD) statement for validation. Results: From an initial sample of 111 patients, 99 were considered for this study. A total of 2177 teeth were examined, with 12 performed extractions. The discrimination of the model was considered good, with an area under the curve value of 0.809 (95% CI: 0.629 - 0.989). Conclusion: Within the limitations of this study, the previously developed predictive model for early tooth loss had a good performance and, therefore, shows validity to be used in this population. Further studies with larger samples and of national representativeness shall be developed to validate this predictive model. Testing the applicability and impact of this model may provide relevant information.
    2020-12Dissertação de mestrado Acesso aberto Ver mais