INTRODUCTION
The root complex of the mandibular first molar (MFM) possesses two roots: a mesial root (MR) and a distal root (DR). However, a third root may be present on the lingual side of the tooth or very rarely on the buccal side of the tooth [30]; the former is named radix entomolaris (RE) and the latter radix paramolaris (RP).
The prevalence of RE shows a distinctive pattern of geographic variation [23]. Also, this root appears to be more frequent in individuals affected by Turner syndrome than in healthy individuals [17]. In North and East Asians and Eskimo-Aleuts, this trait reaches a prevalence of > 20%, which sets them apart from all other human populations. By contrast, it rarely occurs in Western Eurasian, Sub-Saharan African, New Guinean, Melanesian, and Indian populations (< 5%). A recent radiographic study reported a prevalence of 2.36% (95% confidence interval [CI] 1.61–3.42%) for the Slovene population [27]. As such, RE is among the traits used for ancestry estimation in the context of forensic dental profiling. A recently developed web-based application rASUDAS estimates the ancestry of unknown individuals based on their suite of tooth crown and root traits, including RE [22]. In contrast, RP seems to be rare in all human populations [5].
Apart from their role in forensic anthropology, RE and RP represent a diagnostic and therapeutic challenge in clinical dentistry, particularly in oral surgery, endodontics and periodontology. First, these roots are prone to fracture during the extraction of the tooth. Second, they may be the cause of endodontic treatment failure if not identified, properly instrumented and filled. Third, their presence may contribute to localised periodontal destruction, which is probably related to the increased complexity of furcation morphology [13]. As a result of all these factors, three-rooted MFMs are lost in significantly younger patients than their two-rooted counterparts [8].
Knowledge of the morphological characteristics of three-rooted MFMs is essential for improving the success rate of various treatments and reducing the frequency of procedural complications. In the previous decade, cone-beam computed tomography (CBCT), spiral computed tomography (SCT) and micro-computed tomography (micro-CT) have been used to analyse several morphometric aspects of such teeth, including pulp chamber geometry, root canal configuration, root canal curvature and odontometric variables [9, 10, 16, 20, 25]. However, until now, few studies have addressed the morphology of three-rooted MFMs in European populations [4, 5, 16]. Thus, this study aimed to analyse the morphometric aspects of the external and internal anatomy of three-rooted MFMs extracted from dental patients in Slovenia.
MATERIALS AND METHODS
Extracted teeth
The Faculty of Medicine of the University of Ljubljana houses a collection of extracted teeth donated by Slovene dentists. The collection presently includes 17 three-rooted MFMs (11 right and 6 left). All of them have a lingually-located supernumerary root (RE). The teeth and their supernumerary roots were identified by the present authors. For most teeth, there was no data on the patients’ gender, age, ethnic background, and the reasons for extractions.
Macroscopic analysis of the external dental morphology
The following observations and measurements were recorded:
All measurements were made with a digital calliper with a resolution of 0.01 mm (ABS Digimatic, Mitutoyo, Japan).
CBCT analysis of the internal dental morphology
The CBCT images of the extracted teeth were taken with a Veraviewepocs 3D R100 device (Morita, Kyoto, Japan). The operating parameters were 8 mA, 90 kV, field of view 42.6 mm × 43.0 mm, slice thickness 1.0 mm, and voxel size 0.125 mm3. The obtained images were analysed using i-Dixel One Volume Viewer 2.0.0 software (Morita, Kyoto, Japan).
The following observations and measurements were recorded:
Data analysis
All measurements were made independently by both authors, and the average was used for data analysis. A t test was applied to examine any statistically significant difference between the lengths of both main roots and RE, between the levels of mesio-distal and distal furcations, and between the RE canal curvature in CV and PV. A p-value of < 0.05 was considered significant.
Literature search
A literature search for articles on MFMs with RE published until August 2019 was conducted using a PubMed database search. The following keywords were used: ‘three-rooted mandibular first molar’, ‘radix entomolaris’, ‘disto-lingual root’, ‘additional distal root’, and ‘extra distal root’. Papers were retrieved and screened for three-dimensional (3D) pictures of MFMs with RE (photographs of extracted teeth and 3D reconstructed CBCT, SCT and micro-CT images of teeth). Cases of teeth depicted only two-dimensionally (periapical radiographs, CT slices) were excluded. Reference lists were checked to identify any other relevant articles.
On the selected pictures, the following observations were recorded:
All pictures were independently evaluated by both authors. Disagreements were resolved by joint evaluation.
RESULTS
Macroscopic analysis of the external dental morphology
We were able to sort the examined teeth into types A (n = 16) and AC (n = 1) defined by Carlsen and Alexandersen [4]. In type A, the cervical part of the RE was fused with the DR, whereas in type AC the RE was located exactly midway between the MR and DR (Fig. 2). The only tooth with AC morphology was extracted from a 22-year-old female patient of Slovene origin after unsuccessful endodontic treatment. It was undoubtedly an MFM, since all three mandibular molars were present in the tooth row before extraction.
Table 1 shows the distribution of teeth according to the curvature and relative length of the RE [7, 14, 24]. The average levels of the root apexes and root furcations can be seen in Table 2. The RE was significantly (p = 0.001) shorter than the DR, with an average difference of 2.04 mm (range 0.07–3.50 mm). It was also significantly (p < 0.001) shorter than the MR, with an average difference of 3.15 mm (range 1.60–4.95 mm). The level of the distal furcation was significantly (p = 0.003) lower than that of the mesio-distal furcation with an average difference of 1.39 mm (range 0–6.22 mm). All three roots separated at the same level in 4 MFMs (3 of type A and 1 of type AC). In type A, RE’s degree of separation, in relation to the DR, varied from 0.30 to 0.85 with an average value of 0.64.
Reference |
Country |
Method |
N |
Conical type |
Small type |
Type I |
Type II |
Type III |
Song et al. [24] |
Korea |
CBCT |
358 |
5 (1.4%) |
9 (2.5%) |
29 (8.1%) |
170 (47.5%) |
145 (40.5%) |
Kim et al. [14] |
Korea |
CBCT |
507 |
– |
8 (1.6%) |
39 (7.7%) |
141 (27.8%) |
319 (62.9%) |
De Moor et al. [7] |
Belgian |
ET, CE |
18 |
– |
– |
2 (11.1%) |
5 (27.8%) |
11 (61.1%) |
Present study |
Slovenia |
ET |
14 |
0 (0%) |
0 (0%) |
1 (7.1%) |
3 (21.4%) |
10 (71.4%) |
Structure |
Sample size |
Minimum [mm] |
Maximum [mm] |
Mean [mm] |
P |
Mesial root |
15 |
12.24 |
16.75 |
14.38 (14.09) |
< 0.001 |
Distal root |
15 |
9.82 |
14.85 |
12.30 (13.64) |
0.001 |
Radix entomolaris |
12 |
8.60 |
12.87 |
11.10 |
|
Mesio-distal furcation |
16 |
2.02 |
4.85 |
3.37 |
0.003 |
Distal furcation |
16 |
2.60 |
9.59 |
4.77 |
|
CBCT analysis of the internal dental morphology
The RE and DR of the examined MFMs exhibited Vertucci type I root canal system (only one DR with type III canal system was the exception). The MR showed a more variable canal configuration; however, the prevailing anatomy was type IV (10 cases), followed by type II (4 cases). The 3-2-1 canal system was observed in one case. The canal configuration in the remaining two MRs could not be determined.
Divergence of the RE canal, in relation to the DR canal varied from 30.90° to 75.58° with a mean value of 53.14°. All RE canals contained a primary curvature in the buccal direction; in one tooth, there was also a secondary apically-located curvature in the opposite, i.e. lingual, direction. In PV, all RE canals were classified as severely curved. In CV, however, only one severely curved RE canal was found. Statistical analysis showed that the average RE canal curvature was significantly (p < 0.001) greater in PV than in CV (Table 3).
View |
Sample size |
Minimum [o] |
Maximum [o] |
Mean [o] |
P |
Proximal |
8 |
30.71 |
69.61 |
48.48 |
< 0.001 |
Clinical |
8 |
0.00 |
39.10 |
10.04 |
|
The average inter-orifice distance from the RE canal to the DR canal was 3.59 mm (range 2.39–4.50 mm). The RE canal orifice was located disto-lingually to mesio-lingually from the DR canal orifice (Fig. 3).
In the mesio-distal direction, its position varied from 1.44 mm mesially (type AC morphology) to 0.61 mm distally from the DR canal orifice.
Literature search
The literature search identified 29 relevant papers with 3D pictures of 58 MFMs with RE (39 right and 19 left). Type A morphology was identified in 54 MFMs; in three of them, RE was non-separate in relation to the DR [4, 18]. RE shorter than half of the length of the associated DR was observed in one tooth [14]. Type AC morphology was identified in four mandibular first molars [25, 26, 28]; data on these teeth are summarised in Table 4. In three of them, RE was shorter than half of the length of the associated DR.
Tooth (FDI-notation) |
Age [years] |
Gender |
Country |
Ethnicity |
Method |
Relative length |
Reference |
46 |
NS |
Male |
Canada |
Inuit |
ET |
< 1/2 |
Tumer [28] |
46 |
NS |
Male |
Canada |
Inuit |
ET |
< 1/2 |
Tumer [28] |
46 |
NS |
NS |
Senegal |
Wolof |
ET |
< 1/2 |
Sperber and Moreau [26] |
46 |
NS |
NS |
Brazil |
NS |
Micro-CT |
> 1/2 |
Souza-Flamini et al. [25] |
46 |
22 |
Female |
Slovenia |
Caucasian |
ET |
> 1/2 |
Present study |
DISCUSSION
The material available for this study was small (17 MFMs with RE), which is a characteristic problem when studying relatively rare root traits in small ethnic areas, such as Slovenia. This is also due to the decreasing availability of extracted teeth for research purposes, which is associated with the decline in the prevalence of caries and advances in dental care in this country. Nevertheless, the teeth were well preserved, and it was possible to perform most of the observations and measurements. Moreover, in one MFM, the RE was located exactly midway between the MR and DR, corresponding to type AC as defined by Carlsen and Alexandersen [4]. A tooth with such morphology has not been found even in much larger collections of extracted teeth, e.g. in the Copenhagen and Amsterdam collections that together included 87 MFMs with RE [4, 30].
Our finding challenges the widespread belief that RE in MFMs invariably appears disto-lingually, with its coronal portion fixed to the DR (type A variant) [1, 3, 4]. Various positional terms (disto-lingual root, supernumerary disto-lingual root, disto-lingual extra root, additional distal root and extra distal root) have almost completely replaced the more common term ‘RE’ in the recent dental and anthropological literature. According to the results of our literature search, only four MFMs with type AC morphology have been documented (Table 4). Surprisingly, none of the authors mentioned that this was an unusual root morphology of the MFM. In three of these teeth, RE was shorter than half of the length of the distal root. The ethnic origin of individuals with type AC MFMs was variable; however, the present study appears to provide the first description of such a tooth in an ethnic European individual.
A small number of documented MFMs with type AC morphology probably reflects the rarity with which they occur. An additional reason might be that such teeth are discarded by oral surgeons, because it is even less possible to extract them in one piece than their more prevalent type A counterparts. In our case, the tooth was extracted in several fragments that were collected and later glued together and only then the unusual root morphology became apparent. Fortunately, the lingual side of the root trunk was not damaged during extraction, and it was possible to register the position of RE in the root complex with certainty.
In our study, the RE was always longer than half the length of the DR (short and conical types absent) and most frequently (71.4% of the teeth) tilted lingually and curved buccally (type III). A Belgian study of extracted and endodontically treated teeth [8] and a Korean CBCT-based study [15] yielded comparable results; however, another Korean CBCT-based study [26] reported similar frequencies of types II and III (Table 1).
It is often claimed in the literature that the RE is shorter than the MR and DR; however, numerical data have been presented by only a few authors. In the examined MFMs, the RE was on average 2.04 mm shorter than the DR and 3.15 mm shorter than the MR. Both differences were statistically significant (p ≤ 0.001). Our results are in line with studies on three-rooted MFMs extracted from Taiwanese [6] and Chinese patients [10]. The former reported that the RE was significantly shorter than the DR with an average of 1.48 mm. In the later, RE was significantly shorter than both DR and MR with an average of 0.88 mm and 1.54 mm, respectively. A CBCT study in Korea has shown that MFMs with RE shorter than half of the length of the associated DR represent around 4% of all cases [24]. Therefore, in MFMs, the RE mostly appears as a well-formed root approaching the length of both main roots. In the present study, the average lengths of the MR and DR were similar to those reported for two-rooted MFMs extracted from dental patients in Slovenia [15] (Table 2). These findings are of clinical significance when determining working lengths during the endodontic treatment.
Knowledge of the distal furcation level plays a significant role when planning periodontal treatment, especially because it is not possible to detect this furcation with conventional radiography. In four of the examined MFMs (3 type A and 1 type AC), RE separated from the root trunk at the same level as both main roots, forming a trifurcation. These teeth would be suitable for resection of RE in the case of furcation involvement, because a ‘normal’ two-rooted morphology could be established in this manner. In the remaining 13 MFMs, the RE separated from the DR apically from the mesio-distal furcation. Therefore, distal furcation would become periodontally involved later in the disease process when the amount of periodontal tissue support may not be sufficient to allow RE resection. The level of the distal furcation was significantly (p = 0.003) lower than that of the mesio-distal furcation, with an average of 1.39 mm. These results are concordant with those obtained in a Chinese study [10]. Moreover, the RE’s degree of separation, in relation to the DR, varied considerably (from 0.30 to 0.85). Our literature search indicates that, in extreme cases, a complete fusion (degree of separation 0) between the RE and DR may also occur [4, 18].
In Asian populations, the root canal anatomy of the MFMs with and without RE has been comparatively studied in vivo using CBCT [32], and in vitro using micro-CT [9] or the method of canal staining and root clearing [11, 12]. The MR predominantly contained type IV (29–94%) and type II (2–31%) canal configurations in both, the two- and three-rooted molars. In contrast, the DR less frequently contained two canals in the three-rooted molars (5–14%) as in the two-rooted molars (28–36%). These results, together with those from our study, indicate that the occurrence of the RE is associated with a tendency for simplification of the canal system in the DR, but not in the MR.
In all examined MFMs, the RE canal was severely curved (> 25°) bucco-lingually, which is in accordance with previous studies that established a severe bucco-lingual curvature of the RE canal in 90.5% to 100% of the examined MFMs [6, 9, 20]. This pronounced curvature of the RE canal has to be taken into consideration during root canal instrumentation, because it increases the risk of complications such as over-straightening with strip perforations or instrument fractures. Yu et al. [33] showed that a significantly greater incidence of instrument separation, lateral perforation, and underfill was associated with the RE canal than with other root canals in MFMs. Also, our study showed that there was a considerable lingual inclination of the RE canal, ranging from 30.90° to 75.58°. Therefore, the endodontic instruments have to be inserted into the RE canal from the buccal side. Worthy of note is that the two clinically relevant morphometrical characteristics of RE (buccolingual curvature and lingual inclination) cannot be visualised using conventional two-dimensional radiography.
Several authors measured the distance on the pulp floor between the orifices of the DR and RE canals. For teeth extracted from Chinese patients they reported the average inter-orifice distance of 2.93 mm [8], for those extracted from Korean patients of 3.77 mm [14], and for those extracted from Brazilian patients of 3.20 mm [25] and 3.93 mm [20]. A comparable average value of 3.59 mm was determined in the present study. However, in the above-mentioned studies, the position of the RE canal orifice, in relation to the DR canal orifice, has not been investigated, although it has implications for the endodontic access cavity design. We were able to demonstrate the positional continuity between the RE canal orifices located slightly disto-lingually and those located considerably mesio-lingually from the canal in the DR. Based on literature review, Abella et al. [2] suggested that the disto-lingual corner of the access cavity should extend more lingually to facilitate the localisation and access to the RE canal. To this we can add that in MFMs with type AC morphology the extension should be directed mesio-lingually (as seen in Fig. 2).
The radiographic identification of a disto-lingually located RE (type A) may be a challenging task, because this root is mostly situated in the same bucco-lingual plane as the DR. Therefore, the additional 25° mesial radiograph is essential in such cases [31]. In MFMs with mid-lingually located RE (type AC), overlapping with the DR is less likely, which makes it easier to identify by using periapical radiography. The absence of overlapping between the RE and DR appears to be the only sign indicating type AC morphology on a straight periapical radiograph. CBCT overcomes these limitations and allows a reliable distinction between both morphological types of RE.
The absence of MFMs with RP in our material is not unexpected, given the rarity with which they occur. Such a tooth was not found among 1954 MFMs from the Amsterdam collection of extracted teeth [30], nor was it found on CBCT scans of 116 Brazilian dental patients [20]. However, one MFM with a separate RP and four MFMs with unseparated RP and MR were identified in the Copenhagen collection of extracted teeth [5].
CONCLUSIONS
Mid-lingual position of a RE is a rare but clinically significant anatomic variation of MFMs. This variation needs to be identified during endodontic treatment so that appropriate adjustments to the access opening can be made, and thorough debridement of the additional root canal can be accomplished. From a periodontal viewpoint, such a tooth is suitable for resection of the RE in the case of furcation involvement. This study disputes the previous assertion that an RE in MFMs always occurs on the lingual side of the main distal root. Replacement of the more inclusive term ‘RE’ with positional terms stating the distal position of the additional root (disto-lingual/additional distal/extra distal root) is hardly appropriate in light of the presented findings. Regardless of the RE position in the root complex, the clinician should anticipate that its root canal exhibits a lingual tilt and a severe curve toward the buccal, which cannot be detected readily on conventional radiographs.