Background: In patients with periodontitis, a quantitative prognostic assessment is needed in order to make evidence-based decisions about retaining teeth or extracting and replacing them with a dental prosthesis.
Methods: One hundred and ninety eight patients receiving active periodontal treatment in 1989 or 1990 and complying with supportive periodontal therapy (SPT) over an average of 11.8+/-2.3 years were included in the study. A generalized linear model was established and fitted via generalized estimating equations to identify predictors for tooth loss during SPT.
Results: Of the 4559 teeth present at baseline, 166 (3.6%) were extracted during active treatment and 249 (5.5%) during SPT. Baseline findings of diabetes mellitus (OR=4.17), reduced alveolar bone levels (OR=1.04 for each 1% increment), increased tooth mobility (III versus 0: OR=5.52), multiple roots (OR=1.82), and non-vital pulp (OR=2.24) were significant (p<0.05) predictors for tooth loss during SPT. Based on these parameters, a prognostic model was constructed that provides estimates of tooth survival probability when periodontal therapy is rendered.
Conclusion: Using a multivariate approach, a prognostic model was developed that may be of value for clinical decision making.