INTRODUCTION
The main objective of endodontic treatment is to prevent or treat apical periodontitis 1 . The persistence or emergence of an apical periodontitis lesion after treatment can be regarded as a failed outcome and may be associated with difficulties encountered during endodontic intervention 2 .
Chemomechanical preparation can be considered as the main phase of root canal treatment 3 . It consists of cleaning, shaping, and disinfecting the main canal, through the mechanical action of instruments and the chemical effects of irrigant solutions, creating appropriate intracanal conditions to receive the filling material 4-6 .
Anatomical complexities ofthe root canal system can pose a major challenge during endodontic treatment, so knowledge of internal anatomy and its variations is of utmost importance for success 7 , 8 . Endodontic instruments and irrigants have limitations in reaching and disorganizing bacterial biofilms located in areas such as isthmuses, ramifications and recesses, which often require special strategies for cleaning and disinfection 1 .
Difficult-to-reach irregular areas can harbor remnants of pulp tissue or residual infected debris, which may compromise the treatment outcome 9 . Mandibular premolars can present complex anatomy; according to the Vertucci’s classification 7 , these teeth can present several anatomical classifications, from type I to V. In addition to these variations, these teeth can also present a C-shaped canal, which is a ribbonshaped orifice, formed when the canals merge to form a 180° arc and a narrow strip of curved pulp tissue is formed 10 . A C-shaped canals presents a serious challenge for adequate endodontic treatment. Root canal anatomy may vary according to ethnic factors 11 , sex 12 and age 13 . Most previous studies were performed in Caucasian populations 14 . Similar investigations in other populations in South America are less frequent, especially in Colombia, where studies on internal dental anatomy are rare. Thus, the purpose of the present study was to describe the root canal morphology of the mandibular first premolar in a Colombian population using micro-computed tomography (micro-CT) as the evaluation method.
MATERIALS AND METHODS
Fifty mature mandibular first premolars with intact crowns, available from the Bank of Human
Permanent Teeth of the Santo Tomás University, Bucaramanga, Colombia, were used in this study. Teeth had been extracted for orthodontic reasons unrelated to this study. Consent was secured prior to tooth donation. The teeth evaluated in this study were from patients from the metropolitan region of Bucaramanga, Colombia, including the cities Floridablanca, Girón, Lebrija, and Piedecuesta. The population in this region is miscegenated, as in other Latin American countries. This study was conducted under the principles established in Resolution 08430 of Colombia and approved by the Ethics Committee of Santo Tomás University. Exclusion criteria included teeth with incomplete root formation, root resorption, crown and/or root fractures, previous endodontic treatment and extensive restorations and/or caries.
The teeth were scanned on the SkyScan 1174v2 micro-CT device (Bruker-microCT, Kontich, Belgium) with a 50 kV source at 800 μA, with the following parameters: rotation step of 1.0o, a 360° rotation around the vertical axis, and 17 μm pixel size. The image of each specimen was reconstructed from the apex to the cementoenamel junction with the NRecon v.1.6.9 software (Bruker-microCT), which provided transversal axial sections of the internal structure. After this procedure, three-dimensional models of the dentin and canals were obtained through an automatic segmentation threshold with the CTAn V. 1.13 software (Bruker-microCT). Subsequently, the software CTVol v.2.2.1 (Bruker-microCT) was used for visualization and qualitative evaluation of root canal morphology according to the Vertucci classification.
Images of each tooth were evaluated by two observers, and by a third in case of disagreement. The evaluation parameters included: Vertucci’s classification, two-dimensional data of root canal perimeter, circularity, and maj or and minor diameters at 1, 2 and 3 mm short of the apical foramen, as well as three-dimensional data of the total area and volume of the root canals.
RESULTS
According to the Vertucci classification, 40% of the mandibular first premolars were categorized as type I, while 32% were distributed among types III, V and VII (Fig. 1, Table 1). The remaining 28% of the tooth specimens did not meet any of the Vertucci classification types and were considered as additional configurations. Of these, the most frequent configuration was type 1-2-3 (six teeth), followed by configuration 1-3 (three teeth) (Table 2). C-shaped canals were found in 1.8% of the specimens (Fig. 2). Table 3 shows the data from the two and three-dimensional evaluations.
DISCUSSION
Knowledge of the internal and external anatomy of the different tooth groups is essential for an endodontic treatment with favorable prognosis. Some studies have associated the complexity of the root canal system with the failure of endodontic treatment, usually because of the difficulties in attaining proper disinfection throughout the system irregularities 15 , 16 . Thus, the aim of this study was to contribute to the knowledge of the anatomy of mandibular premolars, a tooth with recognized complexity in internal anatomy, from a Colombian population that had not been the subject of an anatomical study by micro-CT.
Micro-CT is an excellent non-invasive, nondestructive tool for assessing internal and external dental anatomy through the reconstruction of threedimensional models. In the present study, a pixel size of 17 μm resolution was used, which is efficient to demonstrate the complexity of root canal anatomy17. In this study, the Vertucci type I configuration was the most prevalent (40% of cases), in contrast to a systematic review that found a frequency of this configuration almost twice as high in other countries 14 .
The prevalence of Vertucci class V canals in the present study of a Colombian population (24%) is similar to prevalence reported in other studies that used micro-CT, including specimens evaluated from populations of China and Saudi Arabia with a frequency of approximately 21% 18 , 19 . It should be noted that some studies18,19 did not evaluate data such as apical canal diameter (Table 3), which could be valuable to plan the instrument size and other preparation strategies 20 .
Vertucci types III and VII configurations had the lowest frequencies in this study, each occurring in 2% of the sample. Similar findings were reported by Liu et al. 21 , who found 2.6% for type III and 0.9% for type VII. Together, types III, V and VII and the additional configurations comprised 60% for the evaluated premolars, highlighting the complex, variable anatomy in this group of teeth. These characteristic configurations in the study population may pose greater difficulties for the clinician during endodontic management 18 , 22 .
The results of this study for canal types III, V and VII are similar to those of another study of internal anatomy using a different methodology, also carried out in Colombia23. However, Vertucci type I was less frequent in the present study. Just over % of the analyzed samples were considered as an additional configuration that did not fit the Vertucci classification, similar to that reported by Alkaabi et al. 19 in an Arab population.
More than half of the evaluated teeth were not Vertucci class I, i.e., they had more than one canal and/or foraminal exit. This condition can be explained by the diversity of the Colombian population and the presence of external radicular grooves, as mentioned in a previous study 9 . These teeth have at least 2 apical foramina, which can be a challenge for proper cleaning and disinfection, increasing the risk for a poor outcome, as the complexity of the root canal system has been associated with the failure of endodontic treatment 4 , 15 .
The Vertucci classification can be a limitation of this study, as it can be considered to be incomplete for studies of internal tooth anatomy. This is because when the classification was created, there was no availability of a method as accurate as micro-CT. The presence of accessory canals does not allow objective classification, and many additional configurations or biases of interpretation may arise in the assessment.
Only one study 22 with the same group of teeth and methodology in Latin America has been carried out to date, so it is not possible to make a broader comparison with other population groups. Further research with micro-CT, analyzing lower premolars in different regions of Latin America, is necessary for better morphometric comparison of this group of teeth.
CONCLUSION
The internal anatomy of mandibular premolars from the studied Colombian population proved to be more complex when compared with some other populations, with most of the teeth not being classified as Vertucci class I. Knowledge of an anatomical pattern can assist the clinician in planning the endodontic treatment and improve the prognosis. More comprehensive studies including more samples should be conducted on this and other populations to gather more data on internal anatomy.