СLASSIFICATION APPROACHES OF ENDODONTIC INSTRUMENTS’ FRACTURES CASES: PROPOSALS FOR THE FORMATION OF PREDICTORS AND PROGNOSTIC MODEL SCHEME REGARDING COMPLICATION DEVELOPMENT

Abstract

Introduction. Literature sources describe approaches for the success prediction of fractured endodontic instrument retrieval, which facilitate the selection of appropriate management strategies for such clinical cases. However, existing models aimed at predicting the risk of occurrence for such complication itself are still primarily based on regression analyses of heterogeneous datasets, which limits the generalizability of the obtained conclusions. Objective of the research. To analyze existing approaches for the classification of endodontic file fracture cases with the consideration of significance levels among key influencing factors, and to identify categories and determinants that could potentially be used as predictors within prognostic model of such complication development. Materials and methods. Selected data was processed using multilevel analytical approach that included: comparative analysis of various classification systems with respect to their discriminatory capacity and prognostic value; grouping process of risk factors according to their origin (clinical conditions, instrument properties and treatment parameters); assessment of potential role of individual variables as regressors in multiparametric prognostic models; analysis of the feasibility for using dynamic reassessment of instrument fracture risk during treatment. Proposed prognostic model for assessing the risk of endodontic instrument fracture was developed through a stepwise analytical process aimed at transforming descriptive and associative factors into formalized prognostic variables suitable for further mathematical and algorithmic modeling, including processing using machine learning computational capabilities. Results and discussion. Provided analysis demonstrated the expediency of shifting from static criteria for instrument wear assessment (such as the number of uses) toward dynamic parameters that reflect actual loading conditions, duration of functional use, and the degree of interaction with dentinal walls under specific anatomical conditions, which may serve as elements of prognostic model. The most consistently reproducible risk factors for endodontic file fracture include the severity and localization of root canal curvature, type and kinematics of instrument motion, rotational speed, geometric characteristics and metallurgical properties of NiTi files, as well as operator experience and specific clinical instrumentation techniques. Generalization of the obtained data allowed to identify three basic groups of determinants for endodontic instrument fracture risk: factors related to initial clinical conditions; factors determined by instrument properties and conditions; and factors associated with the treatment process and the operator. Such structuring approach represents methodologically appropriate basis for constructing a multiparametric prognostic model for this type of complication. Conclusions. The analysis of contemporary scientific sources revealed the absence of unified, prognostically oriented classification of endodontic instrument fracture cases, as most existing approaches are primarily focused on describing fragment localization, while also at selecting complication management strategies rather than estimating the probability of its occurrence. Existing classifications of endodontic instrument damage and deformation (including morphological and kinematic classifications) demonstrate potential for using certain categories as predictors of fracture risk; however, their prognostic value remains insufficiently validated under conditions of multifactorial analysis.