Prevalence of C-shaped mandibular molars in the Portuguese population evaluated by cone-beam computed tomography.

OBJECTIVE: The purpose of this study was to evaluate the prevalence of first and second C-shaped mandibular molars in a Western European population of Portuguese Caucasians.
MATERIALS AND METHODS: Patients having both routine panoramic radiograph and presurgical cone-beam computed tomography (CBCT) exams were selected. The CBCT examination was performed at five different axial levels and the mandibular molars were classified as C-shape according to the Fan criteria. Differences between genders, age groups, left and right side, type of tooth, and root concavity side were determined. The molars were classified as nonfused roots (NFRI-II), FRI-III, or single root with single canal (SS) groups, using the panoramic radiograph. The prevalence of C-shaped anatomy was calculated to each group.
RESULTS: A total of 1783 teeth (695 first molars and 1088 second molars) from 792 patients were included in the study. The prevalence of C-shaped configurations was 0.6% and 8.5% to mandibular first and second molars, respectively. The differences between gender, teeth, and root concavity direction were considered statistically significant (P < 0.05). The FR had a significantly higher rate of C-shapes when compared to NFR (P < 0.05). Intrarater reliability was 93.6%.
CONCLUSIONS: The FR radiographic appearance may be considered a predictor of this anatomy. Second mandibular molars and females had a higher incidence ratio. Apparently, the mandibular C-shaped anatomy is more common in the population of this study than that has been previously reported for the European population.

INTRODUCTION

The C-shaped root canal configuration was first introduced by Cooke and Cox.[1] This anatomic variation has been associated with a possible failure of Hertwig's epithelial root sheath development. Namely, during the embryonic stage of the tooth formation, an anatomy with fused root (FR) canals, complex fins, and interconnecting webs is created.

Even though the C-shaped root canal configuration is a clinical condition, more associated to mandibular second molars,[2] it has also been associated with other groups of teeth.[2] This anatomic characteristic is highly important as is more difficult to manage, namely, to clean, shape, and obturate.

The prevalence of C-shaped mandibular second molar root canal configuration has been studied worldwide.[23] It is well accepted that there is a higher prevalence of this morphology in Asia when compared with other studied regions.[2] However, accordingly to Kato et al.'s systematic review,[2] no study is available in the Western European region. Some authors state that there is an ethnic influence on the prevalence of these configurations.[34] Moreover, in the Caucasian population, there is a lower incidence ratio of C-shaped mandibular molars.[3] Clinical studies analyzing specific ethnic groups are scarce and many of them have several different ethnicities included.[56] The studies regarding the C-shaped mandibular first molars are rare and until now there is no study available in Europe.[5]

The main purpose of this cross-sectional study was to investigate the prevalence of C-shaped mandibular first and second molars anatomy among a specific Western European population, using cone-beam computed tomography (CBCT). The secondary purpose was to search in the panoramic radiographs any feature related to the CBCT that could be a predictor of a C-shaped anatomy.

MATERIALS AND METHODS

Sample population

The Ethical Committee of Implantology Institute in Lisbon, Portugal, approved the research protocol of the present study. CBCT scans and panoramic radiographs, taken from patients receiving treatments, were analyzed and evaluated retrospectively from May 2014 to November 2014. The radiographs were taken for diagnosis purposes and treatment planning. The CBCT scans were performed only before implant surgery procedures in patients that required that specific treatment. Both exams were performed in a Planmeca scanner (Planmeca Promax, Planmeca, Finland) as per instructions of the manufacturer and analyzed using computer visualization software (Planmeca Romexis, Planmeca, Finland). The panoramic radiographs were taken at 68 kV and 13 mA with a 16 s exposure time. The CBCT exam images were obtained at a 0.2 mm voxel size, 80 kV and 15 mA for 12 s and were reconstructed to be 0.6 mm thick slice.

Patients who had both panoramic radiographs and CBCT scans performed between May 2011 and May 2014 were selected for this study. Moreover, only mature first and second mandibular molars detected in both exams were included in the analyzed sample.

Exclusion criteria

Teeth with previous root canal treatment or superimposed artifact images that prevented a correct radiographic evaluation were excluded from the study.

The same evaluator, with a 3 months interval, performed the evaluation of both CBCT and panoramic radiographs. The evaluator was blinded regarding the CBCT and panoramic radiographic correspondence.

Cone-beam computer tomography analysis

The CBCT evaluation was performed by observing the axial cross-section images at five levels of the root canal system: 2 mm under the canal orifice (“coronal”), 2 mm above the anatomic apex (“apical”), middle distance from canal orifice and anatomic apex (“middle”), middle distance between “coronal” and “middle” (1/3), and middle distance between “middle” and “apical” (2/3) [Figure 1]. The teeth were classified as mandibular C-shaped molars accordingly to the classification proposed by Fan et al.[7] Only the teeth that had simultaneously the following three characteristics were classified as C-shaped molars: FR, the presence of a longitudinal groove on lingual or buccal surface of the root and at least one axial cross-section of the canal belonging to the C1, C2, or C3 configuration. The axial cross-section classification used was also the one proposed by Fan et al.,[7] which is a modification of Melton et al.'s method.[8] This classification has five categories: An interrupted C-shaped canal (C1), a discontinuous C-shaped canal with a semicolon shape with the angles α and β having no <60° (C2), two or three separated canals with both angles α and β having <60° (C3), single round or oval canal (C4), and no canal lumen (C5) [Figures 2 and 3]. The Planmeca Romexis “Measure Angle” tool was used to analyze the cases that required the angle measurement.

Figure 1

Axial cross-sections levels analyzed using cone-beam computed tomography

Figure 2

Axial cross-section classification of mandibular molars C-shapes[7]

Figure 3

Angles measurements to differentiate cross-section C2 from C3[7]

Panoramic radiographic analysis

The teeth evaluated by radiograph were classified into six groups: Non-FR (NFR) with two divergent or parallel roots (NFRI), NFR with two convergent roots (NFRII), conical FR with mesial and distal root canals merging at the apical level (FRI), conical FR with both mesial and distal canals following their own independent track to the apex (FRII), conical FR with a distal root canal curving toward a radiolucent line when running to the apex and a longer mesial canal following his own path to the apex (FRIII), and a single root with a single canal (SS) [Figure 4]. The first five categories were presented by Sinanoglu and Helvacioglu-Yigit,[9] which is a modification of Fan radiographic classification,[10] the sixth group was added by our team in this investigation.

Figure 4

Radiographic root configurations[9]

Statistical analysis

All teeth were classified, according to their anatomy, by both CBCT and radiographic analysis. The collected data were introduced in SPSS software (IBM SPSS Statistics for Windows, Version 22.0. Armonk, NY, USA) and analyzed. The primary outcome was the prevalence of C-shape configuration. Predictive variables were gender, left and right side, type of tooth, age group, and root concavity side. The proportion (prevalence) of C-shapes in each radiographic configuration and differences between radiographic groups was calculated and expressed with a 95% confidence interval (CI). The Z-test for proportions was used to analyze the differences in independent groups. To all compared groups, P < 0.05 was considered statistically significant. The first 107 teeth included in the study were analyzed twice by the same observer with 1 month time interval and were used to calibrate the observer and to calculate the intraobserver reliability with the Cohen kappa test.

RESULTS

Study sample

A total of 1783 mandibular molars (695 mandibular first molars and 1088 mandibular second molars) from 792 patients (303 males and 489 females) with a mean age of 51 years were included in this study. It was possible to find 97 C-shaped molars in the analyzed sample. Only the cases found in the second lower molars were taken into consideration in the difference between group analyzes.

The intrarater reliability, calculated with a kappa Cohen coefficient for the first 107 teeth screened in the study and was very high with a value of 93.6% and an asymptotic standard error of ±6.3%.

Prevalence of C-shapes according to gender

Sixteen C-shaped second molars were from males (n = 398) corresponding to a prevalence of 4.0% (2.09–5.93% CI 95%) and 77 were from females (n = 690) corresponding to a prevalence of 11.2% (8.85–13.55% CI 95%). The difference between genders was considered statistically significant at P < 0.05.

Prevalence of C-shapes according to left and right side

From the total of 93 mandibular C-shaped second molars, 42 were from tooth #37 (left side) (n = 550) representing a prevalence of 7.6% (5.39–9.81% CI 95%) and 51 were from tooth #47 (right side) (n = 538) with a prevalence of 9.5% (7.02–11.98% CI 95%). The difference between groups was considered statistically nonsignificant at P < 0.05.

Prevalence of C-shapes according to tooth

Regarding the type of tooth, 4 cases were identified in the first mandibular molars (n = 695) [Figure 5] with a prevalence of 0.6% (0.03–1.17% CI 95), and 93 cases were found in the second mandibular molars (n = 1088) representing a prevalence of 8.5% (6.84–10.16% CI 95%). The difference between teeth was considered statistically significant at P < 0.05.

Figure 5

Some examples of the C-shapes found. (a) Two C-shaped mandibular first molars found. Both present a C1 type at the 1/3 axial cross-section level. Tooth #30 also presents an extra distolingual root, which may be understood as a radix entomolaris. (b) C1 type molar. (c) C2 type molar. (d) C3a type molar. (e) C3b type molar

Prevalence of C-shapes according to age group

The higher prevalence of C-shaped anatomy was found among the 45–54 years group with 11.1% (7.23–14.97% CI 95%) while the lowest rate was found in the 65–74 years group with a prevalence of 5.3% (1.19–9.41% CI 95). No case was found in the 85–94 years group (n = 4). The full prevalence list of each age group is shown in Table 1. Excluding the differences with 85–94 years group, which had an extremely low sample size, the only significant difference was between 45–54 and 65–74 years groups (P < 0.05).

Table 1

Prevalence of C-shaped mandibular second molars according to age group

Prevalence of C-shapes according to root concavity direction

The root concavity was turned toward lingual in 73 cases, representing a prevalence of 6.7% (5.21–8.19% CI 95%) of C-shaped second molars with a lingual groove, and only 20 cases with a groove turned to buccal, which means a prevalence of 1.8% (1.01–2.59% CI 95). The difference between groove sides was considered statistically significant at P < 0.05. If we take into consideration only the C-shaped molars as sample, the groove is turned toward lingual in 78% of the cases.

Types of C-shape configurations on each axial level

When analyzing only the mandibular second molar, the most common type of C-shaped configuration to all axial levels is C3. In the present study, this type represents 38.1% on all analyzed axial levels, followed by C2 with 23.4% and only than C1 with 21.1%. The percentage of each C-shaped type on each axial cross-section level is shown in Table 2.

Table 2

Types of C-shaped configuration to each axial cross-section level on the mandibular second molar

Prevalence of C-shapes according to radiographic configuration

NFR configurations identified in the panoramic radiograph presented a low ratio of C-shaped second molars (data crossed with CBCT results) while all FR presented a higher one. FRI had a prevalence of 50.9% (37.44–64.36% CI 95%), FRII had 42.5% (33.39–51.61% CI 95%), and FRIII had 36.4% (7.97–64.83%). The full prevalence list to each radiographic group is shown in Table 3. Both NFRI and NFRII had a significant difference to all the other radiographic groups (P < 0.05). None of the FR groups (FRI, FRII, and FRIII) had difference between them (P > 0.05).

Table 3

Prevalence of C-shaped mandibular second molars according to radiographic configuration

DISCUSSION

The methodologies used to study the mandibular C-shaped morphologies and prevalence rates have changed over the years. Nakayama and Toda,[11] in the early 1940s, performed tooth sections to analyze these clinical conditions. From there, other methodologies such as clearing,[12] radiographic,[13] Spiral CT,[14] CBCT,[15] and micro-CT (μCT)[7] analysis have been used. Although all methodologies have their own benefits and defects, the only available technique that allows a good resolution of three-dimension images of in vivo studies is the CBCT technology. The μCT and clearing techniques allow a high-detail study on the morphologic characteristics of these teeth; however, since both are ex vivo techniques, they may not be the ideal to perform a prevalence study, a type of research where CBCT may be more recommended. The nominal voxel resolution of the CBCT devices varies from 0.400 mm to 0.076 mm, and according to Scarfe et al.,[16] the minimum voxel size used in endodontics should not exceed the 0.200 mm size. The voxel size used to study the C-shaped molars ranged from 0.125 mm[1517] to 0.250 mm.[918] The 0.200 mm voxel size has already been successfully used to study C-shaped molars in previous investigations[19] and is the size selected in the present study. In this research, the classification as a C-shaped anatomy was made according to the Fan criteria, a rigorous classification based on the axial shape analysis that has been recently used in several studies,[91518] allowing a perfect comparison between the researched populations. Due to the lack of homogeneity between studies, mainly regarding the criteria on defining a “C-shape” and the methodologies employed (radiographs, tooth sections, clearing, CBCT, Spiral CT, and μCT), this comparison between populations is a very difficult task.

However, and independently from the heterogeneity of the studies, the prevalence of C-shaped mandibular second molars varies according to the region of the globe and is apparently higher in Asia. The incidence rate in the Chinese and Korean populations may reach 38.6%[15] and 44.6%[4] although lower rates for both populations have also been presented.[202122] In the Middle East, prevalence of 7.2% and 10.6% for Iranian[23] and Saudi Arabian populations[24] has been reported. In North Africa, the Sudanese population may have an incidence rate of 10.0%.[25] No study, before this investigation, was available in European region using three-dimensional images. The only European prevalence study found in a recent systematic review[2] comes from Greece,[13] in the Southeastern Europe, which presents an incidence rate of 4.8% using solely radiographic analysis. However, it has been stated that root canal anatomy[26] and, in particular, C-shaped configurations are not easy to be seen using this radiographic approach,[2] and the criteria used to define C-shape was very different from the one selected to this study, or to the other available studies in different geographic locations.

Our study suggests the existence of significant differences in the prevalence of C-shapes between genders. These findings are corroborated by other studies even though they do not report statistically significant differences.[1518] No differences have been found between molars in the left and right side of the mandible. These results are also in agreement with a similar work in a Chinese population.[15] Regarding the age groups, no differences were found except when comparing 45–54 and 65–74 age groups. These results are also corroborated by similar studies in Chinese[15] and Turkish[18] populations. The deepest root groove is turned toward lingual in 78% of the cases in the Portuguese population, a finding that is in line with a previous study from Korea.[4]

The most common type of C-shaped configuration in all studied levels was the C3, a finding that is in agreement with a research in the Turkish population.[18]

Regarding the presence of this anatomy in the mandibular first molar, the available information is restricted to two studies, one from South America[19] and another from Asia.[27] Both investigations present the C-shaped morphology in the mandibular first molar as an uncommon occurrence, an idea that is corroborated with our findings in the European region.

The preoperative radiographic analysis has been well documented as an important step to diagnosis and treatment planning. It has been stated that FR on mandibular molars, observed radiographically, may be a strong sign of the presence of a C-shaped anatomy.[91013] Our study seems to confirm this suggestion since all fused groups (FRI, FRII, and FRIII) have a significantly higher prevalence of these morphologies than the NFRI, NFRII. Indeed, the higher limit of the 95% CI was higher than 50% to any of the fused groups regarding the prevalence of C-shapes, which may indicate a considerable relation between radiographic FR and the presence of C-shaped anatomy.

One of the main problems of performing large sample size prevalence studies comparing CBCTs and panoramic radiographs has to expose a large number of patients to radiation and the risks associated. We were able to avoid that by accessing an already patterned existing radiographic database, not having to unnecessary expose any patient.

CONCLUSIONS

Although it has been stated that the prevalence of mandibular C-shaped anatomy in European populations is rare, the results of the present study do not corroborate with previous published data. In a Portuguese population, the overall prevalence of C-shaped configurations on mandibular second molars is 8.5%. There are significant differences (P < 0.05) between teeth and root concavity direction. Females are expected to present a higher ratio of C-shaped mandibular second molars. There is also a higher incidence rate of this anatomy in FR types, which may be seen as a radiographic feature that may predict a C-shaped anatomy.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.