Morphology of the bifid mandibular canal and its relationship with mandibular third molar: A retrospective cone-beam computed tomography study

Abstract

Purpose

To investigate the prevalence and morphological characteristics of bifid mandibular canal (BMC) using cone-beam computed tomography (CBCT) in a Greek population and to explore its relationship with mandibular third molar (MTM), given the potential implications for surgical planning during MTM procedures.

Materials and Methods

This retrospective study included CBCT scans of 793 patients (1329 mandibular sides) obtained between 2016 and 2024 for preoperative assessment of MTM extraction. CBCT scans were evaluated bilaterally in coronal, sagittal, axial, and cross-sectional planes by a panel of three experienced oral and maxillofacial radiologists using a standardized assessment protocol. BMC was classified into five types according to its anatomical course. Associations between BMC presence, canal type, age, gender, interface with the MTM, and canal distance from the tooth were analyzed using chi-square and Fisher exact tests (P < 0.05).

Results

BMC was identified in 123 patients (15.5%) and on 139 mandibular sides (10.5%). The retromolar canal was the predominant type (90.7%), followed by forward (4.3%), dental (3.6%), and trifid canal (1.4%); no buccolingual canal was detected. No statistically significant associations were observed between BMC presence and age or gender. Canal type showed a statistically significant association with its interface with the MTM (P = 0.003), whereas canal distance from the tooth was not significantly associated with canal type.

Conclusion

BMC was relatively common in this Greek population, with retromolar canal predominating. CBCT assessment may provide useful supplementary information during preoperative evaluation of MTM to help identify anatomical variations relevant to surgical management.

Introduction

Mandibular canal is a bilateral anatomical structure of the mandible that extends from the mandibular foramen to the mental foramen and contains the neurovascular bundle.1,2 On radiographic examination, the mandibular canal typically appears as a narrow radiopaque linear structure bordered superiorly and inferiorly by cortical bone; however, these borders are not always distinctly visible.2 The term “bifid” is derived from Latin and denotes a division into two distinct parts.3 In the context of the mandibular canal, a bifid canal represents an anatomical variation in which the canal is split into two separate branches. It has been proposed that bifid and trifid mandibular canals arise from developmental variations during embryogenesis, specifically related to the incomplete fusion of three nerve branches responsible for innervating separate groups of teeth.4 Incomplete fusion of these neural elements may result in the persistence of accessory canals, giving rise to bifid or trifid anatomical configurations.5,6 These developmental variations may contribute to the diverse anatomical configurations observed radiographically, highlighting the importance of accurate identification of accessory mandibular canals through contemporary imaging modalities in clinical practice.

Anatomical variations of the mandibular canal have been documented in the literature according to the number of additional branches they present, bifid and trifid.7,8 These variations can occur unilaterally or bilaterally, with bifid canal more commonly observed unilaterally.9,10 Several studies have classified these anatomical variations according to their configuration and location, using panoramic radiographs11,12 and cone beam computed tomography (CBCT).1,2 Naitoh et al.1 classified the variations of the bifid mandibular canal (BMC) into four types based on the origin and course of the additional canal using CBCT imaging: Type 1, retromolar canal; Type 2, dental canal; Type 3, forward canal; and Type 4, buccolingual canal. Moreover, in the literature, the existence of trifid mandibular canal has been reported in a limited number of cases,13,14 and is included among rare anatomical variations, along with double mandibular foramina.15

The clinical relevance of these anatomical variations lies in their potential to affect surgical procedures involving the mandible, including mandibular third molar extractions, orthognathic surgery, dental implant placement, and grafting procedures.16 Failure to recognize these variations may increase the risk of neurovascular complications, such as nerve injury or neuroma formation.17 CBCT is widely recognized as a valuable modality for assessing these anatomical variations, providing precise visualization of the mandibular canal and its branches, along with high-contrast structures, excellent image quality, reduced radiation dose, and improved cost-effectiveness.18,19 Therefore, accurate identification of a BMC prior to any surgical intervention in the mandibular region is strongly recommended to minimize the risk of complications, including failure of local anesthesia, intraoperative and postoperative paresthesia, hemorrhage, hematoma formation, and inflammation.17 Nevertheless, accessory mandibular canals are often overlooked, unrecognized, or misdiagnosed in clinical practice.20

Furthermore, the presence and morphological patterns of BMC have been investigated across diverse populations, including European21,22 and Asian cohorts,23,24 with reported variations in prevalence among different ethnic groups.25 However, considerable variability in the reported prevalence rates and morphological characteristics of BMC exists across studies. Such inconsistencies may be attributed not only to ethnic anatomical differences, but also to methodological heterogeneity, including differences in imaging modalities, CBCT resolution, diagnostic criteria, and classification systems used for canal identification. In addition, conflicting findings regarding demographic associations, such as age and sex distribution, further highlight the lack of consensus in the current literature.7,21,22,23,24,25 In a systematic review, Valenzuela-Fuenzalida et al.7 documented anatomical variations of the mandibular canal and their clinical implications in dental practice, analyzing 5,326 mandibles from Asian populations, 1,405 from the Americas, and 1,297 from European populations.7 Despite these extensive investigations, no studies to date have specifically evaluated BMC in the Greek population using CBCT.

The reported variability in the prevalence and morphology of BMC across different ethnic populations suggests that anatomical patterns may be population-specific and not universally generalizable.21,22,23,24,25 Such differences may have important implications for surgical planning, radiographic interpretation, and the prevention of neurovascular complications during mandibular procedures.7 Consequently, reliance on data derived from other populations may not fully reflect the anatomical characteristics encountered in Greek patients. In addition, detailed evaluation of the spatial relationship between BMC branches and the mandibular third molar remains limited, despite its direct relevance to surgical risk assessment in this anatomically complex region. The absence of CBCT-based studies in the Greek population therefore represents a significant gap in the current literature, limiting the availability of population-specific anatomical data that could support safer and more predictable clinical practice.

The aim of the present study was to investigate the prevalence and morphological characteristics of BMC using CBCT in a Greek population and to explore its relationship with mandibular third molar, given the potential implications for surgical planning during mandibular third molar procedures.

Materials and Methods

This retrospective study was conducted in accordance with the Declaration of Helsinki, and the protocol was approved by the Dental Association of Attica (Athens, Greece), Ethics, Regulation and Research Committee (Ref. 377).

A total of 812 CBCT scans obtained from a private oral and maxillofacial radiology practice between 2016 and 2024, which had been acquired for the evaluation of mandibular third molars prior to surgical extraction, were initially assessed. Only CBCT scans of patients presenting with at least one mandibular third molar, either unilaterally or bilaterally, were included in the sample. Exclusion criteria comprised CBCT scans of patients presenting with mandibular pathology, a history of mandibular surgery, craniofacial syndromes, and scans with insufficient image quality. Following these criteria, 19 patients were excluded, resulting in a study sample of 793 scans (447 females and 346 males). Mandibular third molars present on the right, left, or both sides were included and assessed independently, yielding a total of 1,329 sides. It should be acknowledged that the retrospective design of the study and the inclusion of CBCT scans obtained specifically for third molar assessment may have introduced potential selection bias. Therefore, the study population may not fully represent the general population, which should be considered when interpreting the findings.

All CBCT scans were performed using Newtom VGI Dental Volumetric Tomograph (QR s.r.l., Verona, Italy). The exposure parameters were 110 kV, 4–9 mA, 3.6 s exposure time, and 0.3 mm voxel size. The field of view (FOV) used was 8 cm × 8 cm, 12 cm × 8 cm, or 15 cm × 12 cm, depending on the indication for each examination. Images were analyzed on a FlexScan MX210 color LCD monitor under low-light conditions using NNT software^® (ver. 7.2 – installation package: 7.2.0). Although different FOVs and minor variations in exposure parameters were used according to clinical indication, all CBCT examinations were performed using the same device and a fixed voxel size (0.3 mm), and were evaluated under standardized viewing conditions by a consensus panel of three oral and maxillofacial radiologists (KEA, EC, DNK), which helped minimize potential variability in the detection of small anatomical structures such as accessory mandibular canals. Nevertheless, this variability should be considered when interpreting the findings, as it may have a limited influence on the detectability of very fine anatomical details.

All mandibular third molars (right, left, or bilateral) and their relationship to the mandibular canal were evaluated in coronal, sagittal, axial, and cross-sectional planes. The BMC was classified into four types according to the classification system proposed by Naitoh et al.,1 with the addition of a fifth category of trifid mandibular canal:14 Type 1 – retromolar canal: the retromolar canal bifurcates from the mandibular canal in the region of the mandibular ramus and initially courses anteriorly, subsequently curving to reach the retromolar area (Fig. 1). Type 2 – dental canal: the dental canal arises from the mandibular canal within the mandibular ramus and follows an anterior course toward the root apex of a molar tooth (Fig. 2). Type 3 – forward canal: A. Forward canal without confluence: this variant bifurcates from the mandibular canal in the mandibular ramus and courses anteriorly to the region of the second molar without rejoining the main canal (Fig. 3A). B. Forward canal with confluence: in this subtype, the forward canal originates from the mandibular canal in the mandibular ramus, courses anteriorly, and subsequently rejoins the main mandibular canal (Figs. 3B-E). Type 4 – buccolingual canal: A. Lingual canal: the lingual canal bifurcates from the mandibular canal in the mandibular ramus, courses lingually, and penetrates the lingual cortical plate (Fig. 4A). B. Buccal canal: the buccal canal arises from the mandibular canal in the mandibular ramus and follows a bucco-inferior course (Fig. 4B). Type 5 – trifid mandibular canal: this variant is characterized by the presence of three distinct branches of the mandibular canal (Fig. 5).

Fig. 1. A. Representative anatomical illustration of a Type 1 canal configuration (retromolar canal). B. Cone-beam computed tomography (CBCT) image shows a Type 1 canal on a reconstructed panoramic view. C. CBCT image of the same patient shows the mandibular canal and bifid mandibular canal (BMC, Type 1). The mandibular canal and BMC are highlighted in orange to distinguish them from surrounding anatomical structures.

Fig. 2. A. Representative anatomical illustration of a Type 2 canal configuration (dental canal). B. Cone-beam computed tomography (CBCT) image shows a Type 2 canal on a reconstructed panoramic view. C and D. CBCT images of the same patient showing the mandibular canal and bifid mandibular canal (BMC, Type 2) (C), and a three-dimensional (3D) reconstruction (D). The mandibular canal and BMC are highlighted in orange to distinguish them from surrounding anatomical structures.

Fig. 3. A and B. Representative anatomical illustration of a Type 3 canal configuration (forward canal) without confluence (A) and with confluence (B). C. Cone-beam computed tomography (CBCT) image shows a Type 3 canal with confluence on a reconstructed panoramic view. D and E. CBCT images of the same patient show the mandibular canal and bifid mandibular canal (BMC, Type 3) (D) and a three-dimensional (3D) reconstruction (E). The mandibular canal and BMC are highlighted in orange to distinguish them from surrounding anatomical structures.

Fig. 4. Representative anatomical illustration of a Type 4 canal configuration (buccolingual canal). A. Buccal bifid mandibular canal (BMC). B. Lingual BMC.

Fig. 5. A. Representative anatomical illustration of a Type 5 canal configuration (trifid canal). B-D. Cone-beam computed tomography (CBCT) images showing two accessory canals on a reconstructed panoramic view (B), the trifid mandibular canal (C), and a three-dimensional (3D) reconstruction (D). The mandibular canal and accessory branches are highlighted in orange to distinguish them from surrounding anatomical structures.

Linear measurements were performed to determine the distance between the BMC and the mandibular third molar in coronal, sagittal, panoramic, axial and cross-sectional planes using the measurement tool integrated into NNT software (ver. 7.2.0). Multiple measurements were performed across the different imaging planes, and the final measurement was selected following consensus evaluation by the observer panel, based on the slice demonstrating the closest anatomical relationship between the BMC and the mandibular third molar along the crown or root of the tooth. For each case, only the minimum measured distance—representing the maximum proximity (the closest anatomical relationship) between the BMC and the mandibular third molar—was recorded (Fig. 6). Linear measurements of the minimum distance along the crown or root between the mandibular third molar and the BMC were classified as follows: 0mm, indicating direct contact (interface), and the following distance ranges: 0.1-1 mm, 1.1-2 mm, 2.1-3 mm, 3.1-4 mm, and 4.1 or more mm (Fig. 6).

Fig. 6. Cone-beam computed tomography reconstructed panoramic image shows linear measurement of the distance between the mandibular third molar and the bifid mandibular canal (BMC) at the point of closest anatomical relationship (crown level) (orange arrows). The measured distance between the crown of the mandibular third molar and the BMC is 4.3 mm.

The statistical analysis of the data was conducted using Stata (ver. 15; StataCorp LLC, College Station, USA). All tests were two-tailed, with P < 0.05 considered statistically significant. Associations between BMC presence and age group or gender were evaluated using the chi-square (X²) test, with Fisher’s exact test applied when expected cell counts were less than five. The relationship between BMC presence and the right and/or left mandibular third molar was assessed using McNemar’s test. Associations of BMC type with interface and distance categories were evaluated using X² or Fisher’s exact tests, as appropriate. The analysis of the present study was primarily descriptive and exploratory, based on univariate testing.

Results

CBCT scans of 793 patients, comprising 1,329 mandibular sides, were analyzed for the presence of BMC. BMC was identified in 123 patients (15.5%; 95% CI: 13.2%–18.2%) and on 139 mandibular sides (10.5%; 95% CI: 8.9%–12.3%), while the remaining patients and sides showed no evidence of BMC (Table 1).

Table 1. Prevalence of bifid mandibular canal by patient and mandibular side [number (%)].

CI: confidence interval

The prevalence of BMC across age groups ranged from 11.6% in patients aged 36–45 years to 18.8% in those aged 46–55 years. No statistically significant association between age group and BMC presence was observed (X² test, P = 0.625) (Table 2). Slightly higher prevalence was observed in the ≤25 years and 46–55 years age groups; however, differences between age categories were not statistically significant, indicating that BMC occurrence was independent of age in this study population.

Table 2. Prevalence of bifid mandibular canal according to age group [number (%)].

P = 0.625 by Chi-square test

The prevalence of BMC was slightly higher in males (17.1%) than females (14.3%); however, this difference was not statistically significant (X², P = 0.350), with no statistically significant association observed between gender and BMC presence (Table 3).

Table 3. Prevalence of bifid mandibular canal according to gender [number (%)].

P = 0.350 by Chi-square test

The study population comprised 793 patients, of whom 123 (15.5%) presented with at least one BMC. Among these, 57 cases (patients) were localized to the right mandible, 50 to the left, and 16 were bilateral. A McNemar’s test was conducted to evaluate lateral prevalence, revealing no statistically significant difference in lateral prevalence (P = 0.55) (Table 4). Consequently, mandibular third molars from both sides were treated as independent observations, resulting in a total sample size of 1,329 sides.

Table 4. Distribution of bifid mandibular canals (BMCs) by laterality (right, left, bilateral).

P = 0.550 by McNemar’s test

Regarding BMC morphology, Type 1 was the most common, accounting for 90.7% of all BMCs. Other types were rare, including Type 3 (4.3%), Type 2 (3.6%), and Type 5 (1.4%), while no Type 4 cases were observed (Table 5).

Table 5. Frequency of various types of bifid mandibular canal [number (%)].

A statistically significant association was observed between the presence of an interface with the mandibular third molar and BMC type classification (P = 0.003; Table 6). Notably, all Type 2 and Type 5 cases exhibited a 0 mm distance from the mandibular third molar, indicating direct contact between BMC and mandibular third molar (Table 6).

Table 6. Association between bifid mandibular canal type and mandibular third molar interface (distance = 0 mm) [number (%)].

P = 0.003 by Chi-square test

In contrast, no significant association was found between the measured distance from the mandibular third molar to the BMC and BMC type (Table 7). Nevertheless, 85 of 139 BMC cases presented in close proximity (distance ≤2 mm) to the mandibular third molars.

Table 7. Association between bifid mandibular canal type and distance from the mandibular third molar [number (%)].

Discussion

Accurate localization and awareness of anatomical variations of the mandibular canal are essential for minimizing complications during mandibular surgical procedures.24 These procedures include surgical extraction of mandibular third molars, placement of dental implants, and sagittal split ramus osteotomies.10,24 Comprehending these anatomical discrepancies is instrumental in mitigating potential adverse outcomes, such as neurovascular injury, substantial hemorrhage, development of traumatic neuromas, paresthesia, and anesthesia.16

In the present study, the prevalence and classification of the BMC were assessed using CBCT scans. BMC was identified in 123 (15.5%) of the 793 patients, with no statistically significant differences observed according to age, gender, or mandibular laterality. However, a statistically significant association was found between the presence of an interface with the mandibular third molar and the BMC type classification. The retromolar canal was the most prevalent BMC type, followed by the forward, dental, and trifid canal types, while no buccolingual canal type was observed.

Since the initial report in 1973,26 scholarly interest in the BMC has increased considerably. Studies investigating the prevalence of BMC have employed various imaging modalities, including panoramic radiography,27 CT,28,29 CBCT,1,30,31 and MRI.32,33 Reported prevalence rates vary according to the imaging modality used, with panoramic radiography showing lower prevalence rates (0.08%–2.6%)11,12,27,34 compared with CBCT-based studies (10%–66%).1,8,10,14,35,36,37,38,39,40 In the present study, the observed prevalence of 15.5% was consistent with several previous CBCT investigations.10,14,37,38 Differences in reported prevalence rates may be related to population characteristics and methodological factors, including voxel size, spatial resolution, FOV selection, and observer experience.

The relationship between age and BMC prevalence remains a topic of debate in literature. While some studies reported no statistically significant age-related differences,10,14 others observed lower prevalence rates in younger3 or older35 age groups. In the present study, no significant differences were observed across age cohorts, consistent with similar CBCT studies.10,14 These discrepancies among studies may be related to population characteristics, study design, and methodological differences. Since the present sample consisted of 793 individuals from the Greek population, further studies with larger cohorts are required to confirm these findings. In addition, the retrospective design and the inclusion of CBCT scans obtained for mandibular third molar assessment may have introduced selection bias and limited the generalizability of the results.

The literature presents conflicting evidence regarding the gender-related prevalence of the BMC. While some studies report higher prevalence in males,10,14,23 and others in females,1,41 in the present study BMC prevalence was slightly higher in males (17.1%) than in females (14.3%), with no statistically significant difference, indicating no gender-related association in this sample. Fu et al.29 suggested that these gender differences likely reflect population-specific characteristics rather than variations in imaging modalities. However, inconsistencies across studies may also be attributed to differences in sample composition, imaging protocols, and diagnostic thresholds.

In previous studies, various classification systems for BMC have been established using panoramic radiography11,12 or CBCT imaging.1,10,11,29,36,38,42 In the present study, Naitoh et al.1 classification was applied, with the addition of the trifid canal type.14 The distribution of BMC types in the present study showed Type 1 (retromolar canal) as the most prevalent (90.7%), followed by Type 3 (4.3%), Type 2 (3.6%), and Type 5 (1.4%), with no Type 4 cases observed. Naitoh et al.1 reported type 3 as the most common (59.6%), followed by Type 1 (29.8%), Type 2 (8.8%), and Type 4 (1.8%). In contrast, Rashsuren et al.14 found Type 1 as the most frequent (71.3%), followed by Type 2 (18.8%), Type 5 (5.8%), Type 3 (4.1%), and no Type 4 cases. Overall, our findings are generally consistent with previous CBCT studies,1,10,14 with the retromolar canal being the predominant subtype.14 The identification of trifid canals, although limited, supports their inclusion in BMC classification. Differences among studies may be attributed to classification application, sample selection, and imaging parameters. Although evaluations were performed by a consensus panel of three experienced oral and maxillofacial radiologists to ensure consistency, the lack of independent assessments should be considered as a potential limitation.

Specific BMC variants, particularly the retromolar and dental canal types, may have clinical relevance in surgical planning.3 Awareness of the dental canal course is important for clinical decision-making, especially in molar extractions and other mandibular procedures. In the present study, no significant association was found between BMC type and distance to the mandibular third molar. However, all dental and trifid canal cases showed a 0 mm distance (P = 0.003), while most retromolar canals were located within 0–2 mm of the mandibular third molar. These findings suggest a close anatomical relationship between certain BMC variants and mandibular third molars, which may be clinically relevant. The retromolar canal, in particular, should be carefully considered during surgical procedures involving impacted third molars. Within the study population, 76 of 126 retromolar canal cases showed a distance of 0–2 mm from the mandibular third molar. These observations underscore the importance of careful preoperative assessment. Minor variations in imaging parameters (e.g., FOV and exposure settings), despite the use of a single CBCT device and fixed voxel size, may have influenced image quality and detectability of small accessory canals. In addition, grouping continuous distance measurements into categories may have reduced analytical sensitivity.

Conversely, when soft tissue anesthesia of the lip and chin is achieved but pulpal anesthesia is absent, anatomical variations should be considered.43 Retromolar and dental canal variants should be taken into account during preoperative planning and implant placement to avoid injury to adjacent neurovascular structures. In this context, CBCT is recommended for accurate anatomical assessment.24 Overall, these findings highlight the clinical importance of identifying retromolar and dental canal variants due to their prevalence and potential implications for surgical and prosthetic procedures.8,24

In conclusion, the prevalence of BMC within the examined subgroup of the Greek population was determined to be 15.5%. Among the identified BMC variants, the retromolar canal represented the predominant type, accounting for 90.7% of all detected cases. No statistically significant associations were observed between the presence of BMC and either age or gender. A statistically significant association was identified between the presence of an interface between the mandibular third molar and the BMC type classification. The present findings highlight the importance of careful radiographic evaluation of bifid and trifid mandibular canals during preoperative assessment. In this context, CBCT may represent a valuable imaging modality for the identification and anatomical characterization of these variations prior to mandibular surgical procedures. However, the findings should be interpreted primarily as descriptive anatomical observations, as the small number of certain BMC subtypes limited robust subgroup and multivariate analyses. Further investigations incorporating larger sample sizes and more detailed analyses of BMC morphological characteristics are warranted to enhance the robustness and generalizability of these findings.