Dentin Thickness of the Danger Zone in the Mesial Roots of the Mandibular Molars: A Cone Beam Computed Tomography Analysis

Background : Mandibular first molars appear to be the most commonly tooth subjected to a root canal treatment, therefore a better understanding of the anatomy critical zones for resistance of this teeth may decrease the treatment’s failure rate. So, this study was conducted to evaluate the dentin thickness of the danger zone in mesial roots of mandibular first molars using cone beam computed tomography in an Iranian population. Methods : In this cross-sectional study, 210 Cone Beam Computed Tomography acquisition of the mandibular first molars were collected from a radiology center in Qazvin. The dentin thickness of the mesial roots (mesiobuccal and mesiolingual canals) was measured from the furcation to 5 mm below. The relationship between the dentin thickness in the danger zone and parameters, like age, gender, placement side, root length, the curvature of the canal, canal type, presence of middle mesial canal, and distance between the orifices of the mesial canals was investigated. Frequency, mean and standard deviation for variables were calculated, and data analysis was done by SPSS using simple and multiple linear regression and Pearson correlation coefficient. Also, two-sample t -test was used to compare mesiobuccal and mesiolingual on two sides. The significant level was also considered at ( p < 0.05). Results : The average minimum thickness of danger zone dentin was found to be 0.885 ± 0.259 mm in the mesiobuccal canal and 0.906 ± 0.206 mm in the mesiolingualcanal. Also, the minimum thickness of dentin in the mesiobuccal and mesiolingual canals in the range of 0 to 1 mm from the furcation was more than in other areas. There was no significant relationship between the minimum dentin thickness of the danger zone with gender, placement side, root length, canal type, and mesial canal entrance distance. But with increasing age, the thickness of dentin in the danger zone in the mesiolingualcanal increased significantly ( p = 0.008). It was also observed that with the increase in the curvature of the canal, the thickness of the dentin in the danger zone in the mesiobuccal canal decreased (moderately curved ( p = 0.008), severely curved ( p = 0.046)). In addition, the thickness of the dentin in the mesiobuccal and mesiolingual canal was less in the samples with the middle mesial canal ( p = 0.047, 0.044). Conclusions : Less dentin thickness in the danger zone in the mesial roots of mesiolingual mandibular first molars was seen in younger patients in mesiolingual canal, with a greater degree of canal curvature in the mesiobuccal canal and teeth with a middle mesial canal. Therefore, it is suggested that large taper instruments should be used with more precision to prevent complications.


Introduction
Mandibular first molars are the foremost posterior teeth to erupt, and with an incidence of 17.0%, they are among the most common teeth that need root canal treatment (RCT) [1]. Successful RCT depends on chemical cleaning and mechanical shaping of the canals walls of the endodontic system. Commonly, inadequate cleaning and shaping of the endodontic system can occur for two reasons: insufficiency of understanding of root canal morphology, and inappropriate identification of canals [2]. Therefore, knowledge of the anatomical and morphological vari-ations of the roots, of the intra-canal endodontic system and dentin thickness is necessary to avoid iatrogenic errors and treatment failures [3]. The thickness of the canal wall is an important factor because any mistake about it may lead to problems and compromise the treatment result [4]. Overpreparation of an area where the inner dentin wall is thin can lead to strip perforation, where a lateral, vertical, oblong perforation occurs and communicates between the internal canal system and the external surrounding periodontal ligament and bone [3,5]. These perforations are related to the areas that "Abou-Rass" first introduced as the 'danger zone' concept in the 1980s [6]. In fact, he declared what clinicians had already reported: often, the mesial canals of lower molars don't assume a central position in the root [7]. 'Danger zones' are described as areas where the dentin of the root wall is consistently thin and becomes more fragile and preferred locations for strip perforation during instrumentation [3,8]. Danger zones are commonly seen in mandibular molars [9], maxillary premolars [10], and mesiobuccal roots of maxillary first molars [3,11,12]. On the other hand, 'safe zones' were described as thicker dentin layers, where there is more tooth structure compared to the danger zone of the root [7,13]. Typically, the danger zone is 4-6 mm under the orifice of the pulp chamber [14] and the minimum distal dentin thickness is 1-2 mm under the furcation. The average dentin thickness in the danger zone in the mesial roots of mandibular molars is from 0.78-1.27 mm [1].
Several techniques have been used to evaluate the dentin thickness of root canals and studies have divided them into two types: clinical and laboratory techniques. Clinical techniques include observation during endodontic treatment and radiography such as micro-CT, Cone Beam Computed Tomography (CBCT), etc. Laboratory techniques include sectioning and observing the pulp chamber under a microscope [5]. Conventional radiographs were not a reliable method for measuring dentin thickness, as they showed greater thicknesses than those indicated above [1]. Micro-CT provides clear details about the root canal. Regardless, Micro-CT has a limited sample size and high radiation, so it cannot be used to scan living humans, which restricts its clinical application [1,12]. The limitations of previous radiographic techniques can be overcome with 3D imaging techniques such as cone beam computed tomography (CBCT). CBCT technology has grown rapidly since it was first used in the 1980s because it provides a highquality, precise, non-invasive 3D image for appropriate information about internal root canal anatomy and dentin thickness [1,[15][16][17][18]. CBCT is a useful tool in clinical examination and morphological evaluation. It can provide threedimensional information in axial, sagittal, and coronal sections and eliminates geometric distortion and anatomical superimposition in conventional radiographs [19,20]. One of the limitations of CBCT is the spatial and contrast resolution lower than that of conventional or digital intraoral radiography. Radiographic artifacts are another problem of CBCT imaging. These artifacts may reduce the diagnostic accuracy of CBCT. Also, the movement of the patient during the scan can negatively affect the quality of the final image [21].
Therefore, this study aimed to evaluate the dentin thickness of the danger zone in mesial roots of mandibular first molars using CBCT.

CBCT Images Selection
In this cross-sectional study (analytical, descriptive), 210 CBCT acquisitions of the mandibular first molar, prescribed for treatment plans including root canal, orthodontics, implants, impacted teeth diagnosis, and trauma, were collected using the sampling method available in the archive of a radiology center in Qazvin. All CBCT images were collected by the ProMax 3D scanner device (Romexis software version 3.8.3, Planmeca, Helsinki, Finland) with a voxel size of 0.15 × 0.15 × 0.15 mm and a field of view of 8 × 8 cm.

Inclusion Criteria
1. CBCT with mandibular first molar teeth at least on one side.
2. Teeth with at least two distinct canals up to the middle third in the mesial root.
3. Mandibular first molar teeth with developed apex and without any root resorption and root fracture.
4. Teeth that don't have any caries or resorption on the root surface, periapical lesions, or any odontogenic or non-odontogenic pathology.
5. Teeth that have not undergone root canal treatment and do not have any root canal fillings.

Exclusion Criteria
1. The presence of artifacts from nearby implants or metal crowns makes the measurement impossible.
2. Teeth with C-Shaped canals or calcification where the root canals cannot be identified.
3. Existence of complex root canal systems morphology.
4. Low quality of CBCT images due to movement of the patient during the scan.

CBCT Images Analysis
First of all, the samples were evaluated by an endodontist and a radiologist using Romexis software (R version 3.8.3, Planmeca, Helsinki, Finland) with 95% interexaminer agreement and border of roots was marked. If the two observers did not agree on the border's location of the root, an experienced oral and maxillofacial radiologist would guide them to reach a single decision. Next, the parameters were measured by a trained senior dental student. All the obtained information was recorded in a pre-prepared form. After recording the demographic information and other required information, the length of the mesial roots of each sample, from the furcation to the apex, was measured in the sagittal axis of the CBCT images. And based on the root length, the teeth were classified into 3 groups: long (>10.0 mm), medium (9.0~10 mm), and short (<9.0 mm) (1) (Fig. 1). Schneider's classification [22] was used to check the curvature of the canal in the middle section of sagittal view. In this method, the angle resulting from the collision of two lines, that is, the line drawn parallel to the longitudinal axis of the coronal part of the canal and the line was drawn at the beginning point of the canal curvature to the apical foramen was used to calculate the curvature of the canal. Based on this, the teeth were divided into 3 groups: straight (10 degrees or less), moderately curved (10 to 20 degrees), and severely curved (20 to 70 degrees) [19,23] (Fig. 2).
To investigate the relationship between the mesial canals based on Vertucci's classification [24], the axial view was used and by moving from the coronal to the apical side, the relationship between the two mesial canals was determined. During the examination, if there is a mid-mesial canal, its information was also recorded. The distance between the orifices of the mesial canals was also measured in the same view, at the cemento-enamel junction level of the tooth, and the distance between the buccal surface of the mesiobuccal (MB) canal and the lingual surface of the mesiolingual (ML) canal was also measured to estimate the distance between the entrances of the mesial canals [25] (Fig. 3).
Then, the minimum dentin thickness of the distal wall of the ML and MB canal was evaluated in the axial view of the CBCT images in sections 0-5 mm below the furcation. The measurement was done in such a way that first a hypothetical line was drawn which is tangent to the external walls of the distal side of the mesial root. Then, the depth of the concavity was measured by drawing a line perpendicular to this line in the deepest depression area on the distal side of the root. This work was done in every millimeter of the root surface in sections 0-5 mm below the furcation. The area with the greatest depth of concavity was used to calculate the minimum dentin thickness. This parameter was drawn and measured perpendicularly by drawing a line from the inner walls of the MB and ML canal to the outer surface of the root (Fig. 4).

Statistical Analysis
The collected data is entered into the SPSS v.23 (version 23, SPSS Inc., Chicago, IL, USA) and R software (version 4.1.1, R Foundation for Statistical Computing, Vienna, Austria). Frequency, the average, and standard deviation for variables were calculated and data analysis was done using simple and multiple linear regression and Pearson correlation coefficient. Furthermore, for comparing MB and ML on two sides used a two-sample t-test. The significant level was also considered at p < 0.05.

Descriptive Statistics Results
This study was based on 210 CBCT to evaluate the relationship between the dentin thickness of the danger zone in the mesial roots of mandibular first molar teeth with the parameters of age, gender, placement side, root length, degree of canal curvature, canal type, presence of middle mesial canal, and the distance between the entrances of the mesial canals was done. The collected data was entered into the SPSS and subjected to statistical analysis (Tables 1,2).
The mean of age of participants was 30.53 ± 10.74 and the gender of 59% (n = 124) was female. Out of 210 CBCT images 51.9% (n = 109) were in the left side.

The Dentin Thickness in the Danger Zone
The minimum thickness of dentin in the danger zone in the MB canal varied from 0.32 to 1.72 mm and its average value was measured as 0.885 ± 0.259 mm. In the ML canal, the danger zone ranged from 0.42 to 1.59 mm with an average of 0.906 ± 0.206 mm.

The Location of the Danger Zone
According to Fig. 5, the danger zone was usually seen in the range of 0 to 1 mm from the furcation more than other defined ranges, and it can be concluded that the location of the danger zone was seen in the range of 0 to 3 mm from the furcation with a probability of 93.4 % (196 samples).

Association between the Dentin Thickness and the Patient's Age
There was no significant relationship between the dentin thickness in the danger zone in the MB canal and the age of the patients (p-value = 0.178, β = 0, 95% CI = (0, 0.01)), but there was a significant relationship between the dentin thickness in the danger zone in the ML canal and the age of the patients (p-value = 0.008, β = 0, 95% CI = (0, 0.01)). In this way, the older, the more thickness of dentin in the danger zone in the ML canal (Fig. 6).

Association between the Dentin Thickness and the Patient's Gender
There was no significant relationship between the dentin thickness of the danger zone in the MB canal (pvalue = 0.59, β = 0.02, 95% CI = (-0.05, 0.09)) and the ML canal (p-value = 0.904, β = 0, 95% CI = (-0.05, 0.06)) with the gender of the patients.  Fig. 6. Association between the dentin thickness in danger zone of the mesiolingual canal and the patient's age.

Association between the Dentin Thickness and the Side
There was no significant difference between the dentin thickness in the danger zone with the right or left side of the tooth in the mesial canals (MB canal (p-value = 0.163) and ML canal (p-value = 0.671)).

Association between the Dentin Thickness and the
Curvature of the Roots A significant relationship was seen between the dentin thickness in the danger zone in the MB canal and the degree of curvature of the canal. As the degree of curvature of the canal increases, the dentin thickness in the danger zone decreases moderate canal curvature (p-value = 0.008, β = -0.24, 95% CI = (-0.42, -0.06)), and severe canal curvature (p-value = 0.046, β = -0.17, 95% CI = (-0.34, 0)). But there was no significant relationship between the thickness of dentin in the danger zone in the ML canal and the degree of curvature of the canal medium canal curvature (p-value = 0.078, β = -0.13, 95% CI = (-0.26, 0.01)), and severe canal curvature (p-value = 0.528, β = -0.04, 95% CI = (-0.17, 0.09)).
Based on the multivariable model, only two variables, age, and degree of canal curvature were statistically significant. Based on the step-by-step algorithm and Akaike's criterion (AIC), the best model was identified, and the results of this model showed that in addition to the two variables of age and degree of canal curvature, the presence of the middle mesial canal also had a predictive effect on the minimum thickness of dentin in the danger zone.

Association between the Dentin Thickness and Presence of the Middle Mesial Canal
A significant relationship was seen between the dentin thickness of the danger zone in the MB canal (p-value = 0.047, β = -0.21, 95% CI = (-0.42, 0)) and the ML canal (p-value = 0.044, β = -0.17, 95% CI = (-0.33, -0.01)) with the presence of the middle mesial canal. So, in the samples that had a middle mesial canal, the dentin thickness was less.

Association between the Dentin Thickness and the Distance between the Orifices of the Mesial Canals
Also, there was no significant relationship between the dentin thickness of the danger zone in the MB canal (pvalue = 0.925, β = 0, 95% CI = (-0.04, 0.05)) and the ML canal (p-value = 0.1, β = 0.03, 95% CI = (-0.01, 0.06)) with the distance between the orifices of the mesial canals.

Discussion
The mandibular first molar is the most expected tooth that undergoes endodontic treatment and is anatomically very challenging [26]. Comprehensive knowledge of tooth anatomy is a key factor influencing the outcome of the treatment, in addition to being aware of the most appropriate method for successful endodontic treatment [5]. On the other hand, awareness of the remaining dentin thickness in the roots, especially in the distal part of the mesial roots, which is described as a danger zone, minimizes but doesn't eliminate the occurrence of strip perforation [26]. During canal shaping, dentin thickness is significantly reduced in the danger zone and is prone to excessive attenuation and subsequent complications [1]. According to the study by Berutty & Fedon, even differences of tenths or hundredths of millimeters can be very important to prevent strip perforation [9].
The present study evaluates the relationship between dentin thickness in the danger zone of the mesial roots of mandibular molars with age, gender, placement side, root length, degree of canal curvature, canal type, presence of middle mesial canal and distance between the orifices of the mesial canals based on CBCT images. Currently, the CBCT technique has become very popular in practices because it allows 3D visualization of anatomical structures and minimizes the overlap of surrounding structures.
Various studies have been conducted to measure the thickness of dentin in the danger zone and the factors affecting it [1,26,27]. But based on the searches, no study has comprehensively examined all the possible influential factors. For example, factors such as the curvature of the canal, the type of canal, and the presence of the middle mesial canal were not investigated in any of the available studies. On the other hand, despite the possibility of the effect of ethnicity on the dentin thickness of the furcation area, the available studies on the Iranian population are very limited. According to these limitations, we aimed to design this study and evaluate the mentioned factors.
Dentin thickness reduction is an important factor in canal instrumentation, because widening the root canal too much can cause events such as perforation. Lim and Stock have determined the ideal value of 200-300 µm as the minimum thickness of the canal wall that should remain after instrumentation to prevent perforation or vertical root fracture [28].
Our results showed that the minimum dentin thickness of the danger zone in the MB canal varied from 0.32 to 1.72 mm, and its average was measured as 0.885 ± 0.259 mm. In the ML canal, the minimum dentin thickness of danger zone ranged from 0.42 to 1.59 mm with an average of 0.906 ± 0.206 mm.
Bryant et al. [29] reported that the average danger zone size for 200 canals was 0.79 mm. Keles et al. [30] also reported the thinnest wall of the MB canal, 1.16 ± 0.20 mm, and the ML canal, 1.19 ± 0.18 mm. According to the study of De-deus et al. [7], the values of the danger zone in the MB canal ranged from 0.67 to 1.93 mm with an average of 1.13 ± 0.21 mm and in the ML canal from 0.77 to 1.89 mm with an average of 0.21 ± 1.10 mm. These results were slightly different from the present study. Because different methods were used to measure the thickness of the root canal wall in the danger zone and different fields of the danger zone or different human ethnicities were selected for studies.
To prevent strip perforation in the danger zone, the selection of high taper NiTi rotary instruments should be cautious and also coronal flaring should be limited and instrumentation should be done along the walls that have thicker dentin and away from the walls where the danger zone is [1].
In the present study, the danger zone is observed in the range of 0 to 1 mm from the furcation more than in other areas, and in general, it can be concluded that the danger zone is located in the range of 0 to 3 mm from the furcation with the probability is 93.4%.
Despite our study, Zhou et al. [1] showed that the minimum dentin thickness of MB and ML canals was 3 to 4 mm below the furcation, and there was no difference between MB and ML canals. This difference between our results and Zhou's results could be related to couple of reasons like the difference in the number of samples as well as the ethnical difference.
In the study of Kessler et al. [14], it was shown that the danger zone was located 4-6 mm below the orifice of the canal. This observed contrast could be related to ethnical dissimilarity, the different locations of the origin of the distance measurement, which in their study was the orifices of the canals.
Association between the minimum dentin thickness and age: The present study declared that there was no significant relationship between the dentin thickness of the danger zone in the MB canal and the age of the patients (p-value = 0.178), but this relationship was significant in the ML canal (p-value = 0.008). In this way, older ages correlated to dentin thickness increment in the danger zone in the ML canal.
Zhou et al. [1] divided the patients into three groups: 18 to 30 years old, 31 to 50 years old, and ≥51 years old, and showed that the minimum dentin thickness of the danger zone in the MB and ML canal increased with age in all groups. This significant relationship was in agreement with the results of the present study in the ML canal, but in the case of the MB canal, this relationship was not significant, which can be due to the determination of age ranges as well as the ethnical differences.
So, it should be noted that age is an important factor that affects the thickness of the distal wall of the mesial roots of mandibular first molars. Mandibular first molars in younger people show larger canals and thinner root canal walls than older people [1].

Association between the minimum dentin thickness and gender:
The results of the present study showed that there was no significant relationship between the dentin thickness of the danger zone in the MB canal (p-value = 0.59) and the ML canal (p-value = 0.904) with the patients' genders.
The results of Zhou's study despite of our results showed that the dentin thickness of the danger zone in the MB and ML canals was greater in men than in women, except in mm 1 and 3 from the furcation in the ML canal [1]. The ethnical differences and the different number of samples can be accounted for as the main reasons for these observed differences.
Association between the minimum dentin thickness and the side: There was no significant relationship between the dentin thickness of the danger zone in the MB canal (pvalue = 0.59) and the ML canal (p-value = 0.904) with the tooth side in the jaw.
The results of Zhou et al.'s study [1] also showed that the dentin thickness of the danger zone in the MB and ML canals had no significant relationship with the right or left tooth side, which was in agreement with the present study.
Association between the minimum dentin thickness and the root lengths: The results of the present study showed that there was no significant relationship between the dentin thickness of the danger zone in the MB canal (p-value > 0.05) and the ML canal (p-value > 0.05) with any of the root length classifications. According to De-deus et al.'s study [31], no significant relationship was also found between the dentin thickness in the danger zone and the root length of the mandibular first molar teeth, and the results of this study were in line with our study.
Furthermore, Zhou et al. [1] showed that the dentin thickness of the danger zone was significantly different between long and short teeth. So short teeth had less dentin thickness. In addition to the difference in the number of samples, the reason for these inconsistent results with our data can be related to the ethnical difference, which is one of the important factors affecting the thickness of the distal wall of the mesial roots of the mandibular first molars.
In the study by Dwivedi et al. [26] and Sauáia et al. [27], it was shown that the distal thickness of the mesial roots of mandibular first molars in longer teeth was thinner compared to shorter teeth. Similarly, the reason for this observed difference can be related to the ethnical difference, the different study method, the smaller number of samples, and the different origin of the length measurement, which was the tip of the cusps in the two mentioned studies, and the furcation in the present study.
Association between the minimum dentin thickness and the degree of curvature: The accumulation of stress in the root should be con-sidered during the treatment. Since it is closely related to the vertical fracture of the root. Concerning the accumulation of stress, the curvature of the canal seems to be more important than the external morphology of the root [1]. In this study as the degree of curvature of the canal increases, the dentin thickness in the danger zone decreases. We determined a significant correlation between the dentin thickness of the danger zone in the MB canal and the moderately curved (p-value = 0.008) and severely curved (pvalue = 0.046) degree of curvature of the canal according to Schneider's classification, but no significant relationship was found between the dentin thickness in the danger zone in the ML canal and the degree of curvature of the canal.
Based on the searches, no similar study was found on the relationship between the minimum dentin thickness in the danger zone and the degree of curvature of the canal.
Association between the minimum dentin thickness and the canal type: In the present study, it was also found that there was no significant relationship between the dentin thickness in the danger zone in the MB canal (p-value = 0.484) and the ML canal (p-value = 0.691) with the canal type. Based on the author's information, no similar study was found on the relationship between the minimum dentin thickness of the danger zone and canal type.
Association between the minimum dentin thickness and middle mesial canal: We also found a significant relationship between the dentin thickness in the danger zone in the MB canal and the ML canal with the presence of the middle mesial canal (p-value = 0.047, 0.044). In the samples where the middle mesial canal was seen, the dentin thickness was less.
Likewise, no similar study was found regarding the relationship between the minimum thickness of dentin in the danger zone and the presence of the middle mesial canal.
Association between the minimum dentin thickness and distance between the orifices of the mesial canals: Our study demonstrated it was no significant relationship between the dentin thickness in the danger zone in the MB canal and the ML canal with the distance between the orifices of the mesial canals.
In contrast to our results, De-deus et al. [31] showed that there was a positive correlation between the dentin thickness of the danger zone and the distance between the orifices of the mesial canals. This means that the more distance between the orifices of the mesial canals, the greater the dentin thickness in the danger zone. The possible reason for this difference can be expressed in the smaller sample size, the different methods of measuring the distance between the orifices, and ethnical differences [31,32].

Limitations
In the present study, the investigations were limited to the first molar teeth in the furcation area, and not the whole root, so it's suggested to design a study in which the second molars with the whole root are also examined.
Other limitations of this study include the sensitivity of the measurement method in determining the exact size and location of the studied indicators, the impossibility of separating the cement thickness from the dentin, the impossibility of using methods such as micro-CBCT as a result of conducting this study in vivo.
In this study, individually the CBCT was used; therefore, it's better to design a study in this direction by using the Micro-CT method.

Conclusions
Less dentin thickness in the danger zone in the mesial roots of mandibular first molars was seen in younger patients in ML canal, with a greater degree of canal curvature in the MB canal and teeth with a mid-mesial canal. Therefore, it is suggested that high taper instruments should be used should be used sparingly to prevent root canal perforation and other complications in younger patients, also teeth with a high degree of canal curvature, and teeth with middle mesial canal, especially in the range of 0 to 1 mm from the furcation.

Availability of Data and Materials
The data that support the findings of this study are available from the corresponding authors.

Author Contributions
Conceptualization-MB and RR; methodology-AZ; software-MR; validation-MR, MV and MT; formal analysis-AA; investigation-AZ and MRAG; resources-AA; data curation-MRAG; writingoriginal draft preparation-MT; writing-review and editing-RR; visualization-AZ; supervision-LT; project administration-LT. All authors contributed to editorial changes in the manuscript. All authors read and approved the final manuscript.

Ethics Approval and Consent to Participate
The study protocol was approved by the ethics committee of Qazvin University of Medical Sciences (IR.QUMS.REC.1400.307). ticle and has no access to information regarding its peer review. Full responsibility for the editorial process for this article was delegated to Marco Tatullo.