Impact of root fillings and posts on the diagnostic ability of three intra-oral digital radiographic systems in detecting vertical root fractures

Aim To evaluate (i) the accuracy of three intra-oral digital systems in detecting vertical root fractures (VRFs); and (ii) the influence of different intracanal conditions (without filling, filled with Gutta-percha and metal or fibreglass post). Methodology Forty single-rooted human teeth were decoronated and the root canals prepared. VRFs were induced using a universal testing machine on 20 teeth. Radiographic images were obtained using the paralleling technique with three different horizontal angulations on conventional E/F film, two storage phosphor plates (Digora Optime and VistaScan) and one sensor system (SnapShot). All teeth were radiographed with all intracanal conditions tested. All images were evaluated by five examiners. After 30days, 25% of the images were reviewed. Areas under ROC curves, sensitivity, specificity, accuracy, positive predictive and negative predictive values were compared by anova. Results The weighted kappa values for intra- and interobserver reproducibility were 0.48-0.74 and 0.57-0.66, respectively. The sensitivity revealed a significant difference between imaging modalities, with higher values for SnapShot (0.53) and VistaScan (0.46), followed by the film (0.41) and Digora (0.38). The area under the ROC curve for SnapShot (0.82) was greater than for the other systems in the teeth with a fibreglass post (VistaScan: 0.70; Digora: 0.64; film: 0.60). Conclusion All systems gave a similar performance for detecting VRFs; all gave a good performance in the absence of intracanal materials, but performed less well in the presence of a metal post or Gutta-percha. In teeth with a fibreglass post, digital systems with higher resolution should be used.