Diagnostic value of cone beam computed tomography for root canal morphology assessment – a micro-CT based comparison

Knowledge of root canal anatomy is the first step in planning for a thorough root canal treatment or retreatment. Besides the main root canal considerations, smaller intricacies, including lateral canals and apical deltas, must be considered due to their possible impact in treatment outcome [5, 19].

The advent of high-resolution 3D image acquisition methods has changed the paradigm in anatomical studies in Endodontics. Micro-CT, a non-destructive high resolution 3D imaging method is now considered and used as the gold-standard for evaluation of internal root canal anatomy [14]. However, clinical application is currently not possible due to its high radiation dose. CBCT also allows a high-resolution 3D image acquisition whilst applying a lower radiation dose, making it applicable in vivo, which has contributed to its elevation in both clinical and laboratorial studies [20]. Several publications, with different study designs, have employed CBCT to conduct an analysis of mandibular molar anatomy, but most lack comparison to a reference that is as close to reality as possible [21]. It is, therefore, and especially from a clinical point of view, of interest to compare and validate CBCT imaging as a diagnostic tool, using Micro-CT as reference [22].

Our investigation revealed a higher prevalence of Vertucci type II and type IV morphologies in mesial roots of extracted mandibular molars, and Vertucci type I in the distal roots of the same specimens, which is in accordance with previous reports [21, 23], and additionally showing a high level of agreement between the two diagnostic tools (85,2%) on their identification. Representative 3D reconstructions of the most prevalent anatomies are shown in Fig. 2 taking advantage of a micro-CT resolution not yet available for CBCT imaging. CBCT has been used in both in vivo and ex vivo studies to assess main root canal morphology and, aligned with this investigation, shows globally a good match with the micro-CT gold-standard imaging, validating this method as reliable for main root canal anatomy classification. Both values of specificity and accuracy were considerably high. However, it is important to point-out that the higher the complexity of the root canal anatomy (Types III, V, others), the lower the values of sensibility obtained, meaning that more intricate root canal morphologies are more easily misdiagnosed in a CBCT exam.

Contrary to main root canal morphology assessment, the sensibility of CBCT to detect lateral canals and apical deltas was low and very low, respectively (Table 2; Fig. 1). The literature available is considerably scarcer when referring to the use of CBCT to detect such types of minor anatomic variations. A previous study on the reliability of CBCTs to detect lateral canals before and after root canal treatment analyzed 43 extracted premolars and concluded that CBCT was an unreliable tool for diagnosis of lateral canals. Despite employing micro-CT in the methodology, the scans were only performed after root canal filling and were not used for the purpose of comparison reference [14]. Another investigation aimed at comparing CBCT to periapical radiographs in identifying apical deltas in extracted premolars, using micro-CT as a reference, concluded that CBCT imaging failed at detecting 65% of existing apical deltas in the analyzed specimens [24]. The quality of the images rendered from a CBCT scan can be influenced by several technical and clinical aspects [14]. Particularly in the field of Endodontics, smaller FOVs and voxel sizes such as the ones selected in the present study are recommended since these allow higher spatial resolution [25]. However, high spatial resolution also results in higher noise in the final images, resulting in higher difficulty to identify finer structures [24]. Clinically, previous studies indicate that lateral and accessory apical foramina – and, necessarily, the lumen of the lateral canals and apical ramifications leading to these portals of exist – can be 3 times more constrict than the main apical foramen [26]. An exact protocol determining an optimal equilibrium between voxel size and noise reduction is still to be determined when it comes to identifying minor anatomical structures [27].

Despite the corroborated value of CBCT in Endodontics, its limitations must be acknowledged and considered in adherence to the as low as reasonably achievable principle (ALARA) [3]. The present investigation validates CBCT as a reliable method for main root canal morphology identification and classification in mandibular molars, while raising concerns regarding its applicability when the aim is identifying minor anatomical variations. Further investigations should focus on how to optimize CBCT image acquisition to more accurately allow identification of these intricacies [28]. Also, despite the large clinical sample employed, there was no record of patient data from which the sample was obtained, nor distinction was made between first, second and third molars, thus limiting extrapolation of results. Another limitation of the present assessment is that the screening of the data was performed by a single operator. Future investigations should include more than one observer, thus allowing inter-rater agreement calculation for an increased value of the methodology. Clinically, and since such intricacies are likely to be left untouched by instrumentation, the focus should be on maximizing the effect of chemical debridement with adjunctive techniques such as ultrasonic activation of irrigants [29].