Background Guided endodontics technique has been introduced for years, but the accuracy in different types of teeth has yet to be assessed. The aim of this study is to evaluate the accuracy of three dimensional (3D)-printed endodontic guides for access cavity preparation in different types of teeth, and to evaluate the predictive ability of angular and linear deviation on canal accessibility ex vivo. Method Eighty-four extracted human teeth were mounted into six jaw models and categorised into three groups: anterior teeth (AT), premolar (P), and molar (M). Preoperative cone beam computed tomography (CBCT) and surface scans were taken and matched using implant planning software. Virtual access cavity planning was performed, and templates were produced using a 3D printer. After access cavities were performed, the canal accessibility was recorded. Postoperative CBCT scans were superimposed in software. Coronal and apical linear deviations and angular deviations were measured and evaluated with nonparametric statistics. The receiver operating characteristic (ROC) curve was used to evaluate the predictive ability of angular and linear deviation for canal accessibility in SPSS v20. Results A total of 117 guided access cavities were created and 23 of them were record as canal inaccessibility, but all canals were accessible after canal negotiation. The average linear deviation for all groups was 0.13 ± 0.21 mm at coronal position, 0.46 ± 0.4 mm at apical position, and 2.8 ± 2.6° in angular deviation. At the coronal position, the linear deviations of the AT and P groups were significantly lower than M group deviation (P < 0.05), but no statistically significant difference between AT group and P group. The same results were found in linear deviation at the apical position and in angular deviation. The area under the ROC curve was 0.975 in angular deviation, 0.562 in linear deviation at the coronal position, and 0.786 at the apical position. Statistical significance was noted in linear deviation at the apical position and in angular deviation (P < 0.001). Conclusions In conclusion, this study demonstrated that the accuracy of access cavity preparation with 3D-printed endodontic guides was acceptable. The linear and angular deviations in the M group were significantly higher than those in the other groups, which might be caused by the interference of the opposite teeth. Angular deviation best discriminated the canal access ability of guided access cavity preparation. Graphical Abstract

No abstract

Background Guided endodontics technique has been introduced for years, but the accuracy in different types of teeth has yet to be assessed. The aim of this study is to evaluate the accuracy of three dimensional (3D)-printed endodontic guides for access cavity preparation in different types of teeth, and to evaluate the predictive ability of angular and linear deviation on canal accessibility ex vivo. Method Eighty-four extracted human teeth were mounted into six jaw models and categorised into three groups: anterior teeth (AT), premolar (P), and molar (M). Preoperative cone beam computed tomography (CBCT) and surface scans were taken and matched using implant planning software. Virtual access cavity planning was performed, and templates were produced using a 3D printer. After access cavities were performed, the canal accessibility was recorded. Postoperative CBCT scans were superimposed in software. Coronal and apical linear deviations and angular deviations were measured and evaluated with nonparametric statistics. The receiver operating characteristic (ROC) curve was used to evaluate the predictive ability of angular and linear deviation for canal accessibility in SPSS v20. Results A total of 117 guided access cavities were created and 23 of them were record as canal inaccessibility, but all canals were accessible after canal negotiation. The average linear deviation for all groups was 0.13 ± 0.21 mm at coronal position, 0.46 ± 0.4 mm at apical position, and 2.8 ± 2.6° in angular deviation. At the coronal position, the linear deviations of the AT and P groups were significantly lower than M group deviation (P < 0.05), but no statistically significant difference between AT group and P group. The same results were found in linear deviation at the apical position and in angular deviation. The area under the ROC curve was 0.975 in angular deviation, 0.562 in linear deviation at the coronal position, and 0.786 at the apical position. Statistical significance was noted in linear deviation at the apical position and in angular deviation (P < 0.001). Conclusions In conclusion, this study demonstrated that the accuracy of access cavity preparation with 3D-printed endodontic guides was acceptable. The linear and angular deviations in the M group were significantly higher than those in the other groups, which might be caused by the interference of the opposite teeth. Angular deviation best discriminated the canal access ability of guided access cavity preparation.