Aim: There are currently no prospective studies evaluating the long-term outcomes of non-surgical root canal treatments beyond 5 years, both in terms of treatment success and tooth preservation, and identifying factors predictive of treatment outcome. The aim of the present work was therefore to fill this gap by assessing these outcomes over time and identifying predictive variables based on systematic data collection over a 25-year period.

Methodology: Data concerning the treatments (N = 2500) were systematically collected since 1990. Information was recorded among clinical, technical, radiographic and patient-related characteristics, i.e., approximately 150 variables for each treatment. The data were analysed regarding both treatment success and tooth preservation by multivariable Cox proportional hazards model, and survival curves were generated. The statistical significance level was set at 0.0125.

Results: In total, 56.4% of the treatments could be followed over time (0-25 years, mean = 6.5 years, median = 5 years). Survival probability decreased almost linearly for treatment success, with about 85% after 5 years and 60% after 20 years, and for tooth preservation, with about 90% at 5 years and 50% at 20 years. The variables significantly associated with treatment failure were: pre-operative pain (Hazard Ratio-HR = 1.56 [95% CI 1.23-1.97]), persistent pain (HR = 2.63 [95% CI 1.44-4.80]), good operator rating of treatment prognosis (HR = 0.46 [95% CI 0.36-0.58]), size of periapical bone radiolucency (HR = 1.88 [95% CI 1.67-2.11]), and tooth type (p = .0006). For tooth extraction, they were: combined endodontic-periodontal lesion (HR = 3.37 [95% CI 1.88-6.05]), pre-existing complication before treatment (HR = 1.67 [95% CI 1.26-2.21]), good operator rating of treatment prognosis (HR = 0.45 [95% CI 0.33-0.60]), clinical failure of root canal treatment (HR = 2.78 [95% CI 1.98-3.89]) and tooth type (p = .0012).

Conclusion: Root canal treatment success and tooth preservation on the arch are not static outcomes, but evolve with time. Among a substantial set of potential predictors, only a small proportion was significantly predictive of treatment success and tooth preservation, most of them being disease and patient characteristics, and not technical aspects, except pre-existing complications. These observations challenge the importance frequently given to byzantine considerations related to the numerous technical details of endodontic procedures, as opposed to general concepts of good clinical practice.