The authors have declared that no competing interests exist.
In endodontic practice, clinicians should be aware of possible root canal anatomic variations. The aim of this study was to assess using CBCT acquisitions regarding whether one root canal anatomy of a tooth is associated with a specific anatomy of another tooth.
A total of 106 CBCT acquisitions were obtained using a CBCT scanner with 200μm voxel size. Numbers of roots and canals of the entire dentition were described. Bivariate analyses and logistic regressions were conducted to explore root canal anatomy on one tooth according to age, gender, jaw, side and the others teeth. Multiple correspondence analysis (MCA) was performed to correlate the different numbers of canals profiles.
A total of 2424 teeth were analyzed. Independently from the other variables, the presence of an additional root canal on a mandibular incisor increases the risk of having an additional root canal on a mandibular premolar (OR [95%] 3.7 [1.0;13.2]). The mandibular molar variability increases in women compared to men (OR [95%] 0.4 [0.1; 0.9]). MCA showed correspondence between 2-canals maxillary incisor and canines and 5-canals maxillary molars, and some correlation between additional canal on maxillary and mandibular premolars.
Although CBCT examinations are conducted in the first intention of making a diagnosis or prognostic evaluation, medium FOV acquisitions could be used as an initial database thus furnishing preliminary evaluations and information. In endodontic practice, clinicians should be aware of possible root canal anatomic variations. The visualization of all canals is considered essential in endodontic therapy. The use of multi-correspondence analysis for statistics in endodontic research is a new approach as a prognostic tool.
Root canal systems have been described as complex anatomical structures with significant implication on root canal preparation. In the literature, various factors such as genetics and ethnic differences, have been reported to influence root canal anatomy [
Several studies have investigated root canal anatomy with different
Conventional radiography yields limited information due to the projection in two dimensions (2D) of a 3-dimensional (3D) anatomical structure, which can lead to failure to recognize a root canal because another structure is superimposed upon it [
Considering that the incidence of missed roots or canals in teeth that needed retreatment may be as high as 42% [
Therefore, the aim of this pilot study was to assess using CBCT acquisitions regarding whether one root canal anatomy of a tooth is associated with a specific anatomy of another tooth.
We examined 106 CBCT acquisitions obtained from 2012 to 2013 in a private practice of Oral and Maxillo-Facial Radiology (E. Coudrais EC) in Toulouse, France; all scans were de-identified before being transmitted for analysis (EC). These acquisitions were randomly selected. Authors were not the treating dentists of these patients. Exclusion criteria were patients who were minors, edentulous patients and presence of metallic artifacts. In accordance with ethical and local radiation protection guidelines and the local board, no CBCT examinations were prescribed especially for the study; they were part of routine care. According to French law, the results of medical imaging examinations may be used retrospectively without the patient’s consent when these examinations have been carried out for clinical purposes and when they have been recorded anonymously (article 40–1, law 94–548 of 1 July 1994)[
The CBCT images were obtained using a CBCT scanner (CS 9500 3D®, Carestream, Marne-la-Vallée, France) with tube voltage of 90kV and tube current of 10mA. The voxel size was 200μm and the FOV was 90 x 150mm. The exposure time was 10.8s with a dose—area product of 605 mGy.cm2. The scans were acquired by an experienced radiologist (EC) according to the manufacturer’s recommended protocol with the minimum exposure necessary for adequate image quality.
CBCT scans were analyzed using CS Dental Imaging Software 3D Module v3.2.9 (Carestream, Marne-la-Vallée, France). Two dental practitioners independently (B. Arcaute BA and E. Maury EM) evaluated the images twice at an interval of 2 weeks. A preview of the three planes was conducted and following by a visualization of all slices in the axial plan (coronal to apical and apical to coronal directions). The following variables were recorded: age, sex, type of tooth and number of teeth, jaw (maxillary/mandibular) or side (left/right). For each tooth, the number of roots and canals were noted. When reading the scans, third molars and root fragments were excluded. Have also not been considered teeth with orthodontic appliances, stirred images or presence of metallic artifacts preventing the adequate visualization of endodontic structures.
Statistical analysis was performed using the STATA 13.1 software (StataCorp, TX, USA). Inter and intra-examiner agreement was evaluated by the Kappa test. A random sample of 30 CBCT examinations was re-examined one week after the initial examination to test intra-examiner agreement with the Kappa test. The Kappa test was also used to assess inter-examiner agreement, again using a random sample of 30 CBCT examinations. The scans were read by 2 trained observers (BA and EM). The data were analyzed using descriptive statistics. A Stuart-Maxwell test was performed to compare left/right pairs of teeth (paired samples). When a 2x2 comparison was performed, this test acted as the McNemar test. The level of significance was set at 5% (p<0.05).
The effects of age, gender, jaw (maxillary/mandibular) or side (left/right) on the number of roots and canals were analyzed. To achieve this goal, teeth were grouped (maxillary incisors or canines, mandibular incisors or canines, maxillary premolars, mandibular premolars, maxillary molars, mandibular molars); for each patient, a tooth group was considered with a difference compared to the reference group if at least one tooth of the group had a difference from the average dental anatomy.
Non-parametric and unmatched test (Mann-Whitney) was used to determine the effect of age, non-parametric and matched test (Wilcoxon) for the effect of jaw and side, and chi square test for the effect of gender, on variability. The dependent variable of a root anatomical variation for a specific tooth group was analyzed using a logistic regression model, taking into account the other tooth groups, age and gender. Multiple correspondence analysis (MCA) was performed to correlate different numbers of canals profiles.
From the initial sample of 106 CBCT scans, 102 scans were selected. Four scans were excluded because of presence of a metallic artifact. Patient age ranged from 20 to 88 years, with a mean age of 46.8 years. The study sample included 53 women and 49 men. The inter-examiner agreement of 30 CBCT scans was very high with a Kappa coefficient of 0.86 for root numbers and 0.82 for canal numbers. The intra-examiner agreement of 30 CBCT scans was also very high with a Kappa coefficient of 0.97 for both root and canal numbers.
A total of 2424 teeth were analyzed (1199 maxillary teeth and 1225 mandibular teeth). There was no significant difference between men and women regarding their age or the number of teeth in each dental arch (
Number of patients | Number of teeth | Age | |
---|---|---|---|
N (%) | Mean ± SD | Mean ± SD (min–max) | |
102 | 25.2 ± 4.3 | 46.8 ± 15.9 (20–88) | |
Men | 49 (48%) | 25.5 ± 4.6 | 45.5 ± 15.9 (20–81) |
Women | 53 (52%) | 24.9 ± 4.0 | 48.0 ± 15.9 (23–88) |
We detected no significant different between men and women regarding the age or the number of teeth on dental arch.
ISO | UNS | n | 1 canal 1root | 2 canals 1root | 2 canals 2 roots | 3 canals 1root | 3 canals 2 roots | 3 canals 3 roots | 3 canals 4 roots | 4 canals 2 roots | 4 canals 3 roots | 4 canals 4 roots | 5 canals 3 roots | 5 canals 4 roots |
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
97 | - | 1 (1%) | - | - | - | - | - | - | - | - | - | |||
95 | - | 1 (1%) | - | - | - | - | - | - | - | - | - | |||
96 | 11 (11%) | 1 (1%) | - | - | - | - | - | - | - | - | - | |||
96 | 12 (13%) | 1 (1%) | - | - | - | - | - | - | - | - | - | |||
96 | - | 1 (1%) | - | - | - | - | - | - | - | - | - | |||
95 | - | 1 (1%) | - | - | - | - | - | - | - | - | - | |||
100 | 12 (12%) | 1 (1%) | - | - | - | - | - | - | - | - | - | |||
99 | 13 (13%) | 1 (1%) | - | - | - | - | - | - | - | - | - | |||
94 | - | 1 (1%) | - | - | - | - | - | - | - | - | - | |||
94 | - | 1 (1%) | - | - | - | - | - | - | - | - | - | |||
101 | 3 (3%) | 2 (2%) | - | - | - | - | - | - | - | - | - | |||
101 | 1 (1%) | 5 (5%) | - | - | - | - | - | - | - | - | - | |||
81 | 9 (11%) | 2 (3%) | - | 1 (1%) | 3 (4%) | - | - | - | - | - | - | |||
84 | 14 (17%) | 1 (1%) | - | 1 (1%) | 4 (5%) | - | - | - | - | - | - | |||
95 | 7 (7%) | 6 (6%) | - | - | - | - | - | - | - | - | - | |||
92 | 6 (7%) | 5 (5%) | - | - | - | - | - | - | - | - | - | |||
73 | 4 (6%) | 19 (26%) | - | - | 1 (1%) | - | - | - | - | - | - | |||
74 | 3 (4%) | 20 (27%) | - | - | 2 (3%) | - | - | - | - | - | - | |||
79 | 3 (4%) | 1 (1%) | - | - | - | - | - | - | - | - | - | |||
80 | 1 (1%) | 1 (1%) | - | - | - | - | - | - | - | - | - | |||
78 | - | - | 1 (1%) | - | 1 (1%) | 37 (48%) | - | 1 (1%) | - | - | 1 (1%) | |||
71 | - | - | - | - | - | 30 (42%) | - | - | - | - | 1 (2%) | |||
67 | - | 2 (3%) | - | - | 1 (2%) | 1 (2%) | 19 (28%) | 2 (3%) | - | - | - | |||
63 | - | - | - | - | 3 (5%) | - | 10 (16%) | 4 (6%) | - | - | - | |||
84 | 1 (1%) | - | - | 3 (4%) | 3 (4%) | - | 1 (1%) | 19 (23%) | 2 (2%) | - | - | |||
83 | 1 (1%) | - | - | 5 (6%) | 2 (3%) | - | - | 25 (30%) | 1 (1%) | - | 1 (1%) | |||
75 | - | - | 1 (1%) | - | 5 (7%) | - | 1 (1%) | 2 (3%) | - | - | - | |||
81 | 2 (2%) | - | 3 (4%) | - | 8 (10%) | - | 3 (4%) | 2 (2%) | - | 1 (1%) | - |
The group in underline italics represented the group taken as a reference. We used the ISO and Universal Numbering System (UNS).
Morphological examination of the maxillary and mandibular incisor-canine group showed that these teeth generally had one root for one canal (
(A) Sagittal CBCT image of a mandibular central incisor with one root and two canals. (B) Presence of an additional canal bilaterally in first maxillary premolars (white arrows). (C) Additional canal (white arrows) concomitantly on mandibular incisors, canines and first mandibular premolars, together with no variability on mandibular molars (3 canals, 2 roots).
The first maxillary premolar generally had 2 canals for 2 roots but 3 canals for 3 roots were also found (
The most common morphology for maxillary molars was 3 roots for both first and second molars (
Variability group (difference compared to reference group) | Reference group | |
---|---|---|
1 (1%) | 100 (99%) | |
25 (25%) | 77 (75%) | |
53 (54%) | 45 (46%) | |
18 (18%) | 83 (82%) | |
77 (80%) | 19 (20%) | |
39 (41%) | 57 (59%) |
An age difference was observed for mandibular premolars: reference group was significantly more aged (48.4 ± 1.7 versus 39.5 ± 3.2 years, p = 0.04).
The proportion of women was lower in the variability group compared to the reference group for maxillary premolars (43% versus 62%, p = 0.05), whereas higher for mandibular molars (62% versus 42%, p = 0.05). Globally, both the mean number of roots and canals were lower for maxillary premolars and greater for mandibular molars, when women were compared to men (
An increased variability was observed for mandibular incisors and canines compared to maxillary (p<0.001). The number of canals for incisors, and the number of roots and canals for canines were increased (
The Stuart-Maxwell test did not reveal any significant difference between the right and left side in the present study. However, there appeared to be a trend to more first mandibular molars with 2 roots and 4 canals on the right side (p = 0.06).
Dependent variable | ||||||
---|---|---|---|---|---|---|
OR[95%] | Maxillary incisors/canines | Mandibular incisors/canines | Maxillary premolars | Mandibular premolars | Maxillary molars | Mandibular molars |
Maxillary incisors/canines | - | - | - | - | - | - |
Mandibular incisors/canines | - | - | 0.7 [0.2;1.9] | 1.8 [0.4;7.8] | 1.7 [0.5;4.8] | |
Maxillary premolars | - | 0.6 [0.2;1.8] | - | 2.2 [0.6;7.8] | 0.5 [0.1;1.5] | 1.4 [0.5;3.6] |
Mandibular premolars | - | 2 [0.6;6.8] | - | 4.3 [0.5;38.8] | ||
Maxillary molars | - | 2 [0.5;8.7] | 0.5 [0.1;1.5] | 4 [0.5;42.9] | - | 0.7 [0.2;2.1] |
Mandibular molars | - | 1.7 [0.6;4.9] | 1.4 [0.5;3.6] | 0.3 [0.1;1.1] | 0.7 [0.2;2.2] | - |
Age | - | 1 [0.9;1.1] | 1 [0.9;1.1] | |||
Sex | - | 2.6 [0.8;7.6] | 2.2 [0.8;5.4] | 0.6 [0.2;2.3] | 0.7 [0.2;2.1] |
The odds ratio (OR = eβ) for each covariate in underline italics represented significant results at 95% confidence interval (OR[95%]).
Taking into account the other tooth groups, age and gender, having variability on a mandibular premolar increased 4 times the risk to have variability on mandibular incisors or canines (OR [95%] 4.3[1.2; 15.4]). Conversely, having variability on mandibular incisors or canines increased 4 times the risk to have variability on mandibular premolars (OR [95%] 3.7 [1.0;13.2]). Having variability on mandibular premolars decreased the risk to discover variability on mandibular molars (OR [95%] 0.2 [0.1;0.9]). Finally, being a male decreased about twice the risk to have a variability on mandibular molars compared to being a female (OR [95%] 0.4 [0.1;0.9]). While on some occasions significant, the factor “age” overall contributed minimally to models.
The first dimension of MCA always explained at least 67% of the principal inertia, while the second dimension explained only 12%, and the following dimensions less than 3% (
When the tooth was absent, the number of canals was considered as a new category “s” to be included in the 102 observations analysis of the MCA.
11: 2-canals | |
12: 2-canals | |
13: 2-canals | |
21: 2-canals | |
22: 2-canals | |
23: 2-canals | |
15: 3-canals | |
16: 5-canals | |
26: 5-canals | |
27: 5-canals | |
25: 3-canals | |
45: 2-canals | |
14: 3-canals | |
24: 3-canals | |
33: 2-canals | |
43: 2-canals | |
35: 2-canals | |
37: 4-canals | |
47: 4-canals | |
17: 4-canals | |
31: 2-canals | |
32: 2-canals | |
34: 2-canals | |
41: 2-canals | |
42: 2-canals | |
44: 2-canals | |
36: 4-canals | |
46: 4-canals |
The visualization of all canals is considered essential in endodontic therapy. CBCT images are known to provide three-dimensional information about the presence of additional canals (
Using CBCT acquisitions of 90x150mm FOV, entire upper and lower dental arches can be visualized. Consequently, intra-individual dental variability, the relationship between the additional canals found among the different groups of teeth, may be analyzed.
Consequently, this study will bring elements for the clinician, allowing it to anticipate the degree of technical difficulty and any radiological investigation based on anatomical variations already discovered on other teeth. For example, discovering two canals in a mandibular incisor multiplied by 4 the risk to discover two canals in a mandibular premolar, and conversely.
In the maxillary incisor-canine group, only slightly over 1% of teeth had more than one root or canal. These results are consistent with the literature [
Several authors have studied maxillary molar anatomy. The presence of 3 separate roots was the most common configuration of the maxillary first and second molars [
The morphology most commonly observed for maxillary second molars was 3 roots, with one canal in each root (left 58%, right 65%). The presence of a fourth canal was recorded in 23% (right) to 30% (left), which was similar to previous studies [
The second part of our study objectives was to conduct logistic regression analyses to obtain the risk to discover additional canal on a group of teeth when a variability was found on another group of teeth. To the best of our knowledge, few investigations have looked for left-right symmetry in pairs of opposite teeth, and when such studies have been performed, they have only considered one type of tooth and not the entire dentition [
Root canal anatomy studies with CBCT have described a single type of tooth [
Even if a probabilistic model has been reported in endodontic research to obtain probability of an additional root in mandibular molars, the use of MCA in statistical analyses is a new and original approach. This correspondence analysis model enabled us to form four groups with different number of canals.
The visualization of all canals is considered relevant in endodontic therapy. Can the presence of an additional canal on one tooth predict the chance to discover an additional canal on another tooth in the same patient? Although CBCT examinations are conducted in the first intention of making a diagnosis or prognostic evaluation, medium FOV acquisitions could be used as an initial database thus furnishing preliminary evaluations and information. Multi-institutional studies may be undertaken to collect more CBCT images and perform additional statistical analyses.
(DOCX)
(DOCX)
We thank E. Coudrais for the CBCT database and N. Crowte for language revision. We acknowledge the statistical advice provided by J.N. Vergnes and G. Fernandez de Grado. We also acknowledge the research platform of the Toulouse Dental Faculty.