Assessment of the Anterior Loop of the Mental Nerve Using Cone Beam Computerized Tomography Scan

Surgery in the anterior mandible such as implant placement in the interforaminal area or chin grafting may damage the anterior loop of the mental nerve, resulting in neurosensory disturbances. During surgery, surgeons usually expose the mental foramen to visualize the position of the mental nerve. However, without knowing the anterior loop length, surgeons have a high risk of violating the anterior loop, when present.

Preoperative radiographic examination aids in developing a comprehensive treatment plan for patients who need dental implant surgery. It helps determine the proper size, location, and angulation for each dental implant as well as the number of implants to be placed.^1,2  Some authors claim that 2-dimensional (2D) radiographs are sufficient for presurgical implant planning³ ; however, others believe there is a need for additional cross-sectional imaging.^4–6  Recently, it has been recommended that cross-sectional imaging be used for the assessment of all dental implant sites and that cone-beam computerized tomography (CT) is the imaging method of choice for gaining this information.⁴ 

When planning to place implants in the mandible, one of the most important anatomic landmarks is the mandibular canal, in which the inferior alveolar nerve passes through and continues anteriorly as the mental nerve, exiting at the mental foramen. The mental nerve may have an anterior loop, which is described in Sicher's Oral Anatomy.⁷  The anterior loop is described as “the mental canal which rises from the mandibular canal and runs outward, upward and backward to open at the mental foramen.” The location of the most distal implants in the interforaminal area when planning an implant-supported fixed complete denture is determined by the position of the mental foramen and the anterior loop of the mental neurovascular bundle.^8–10  To place the implant closest to the mental foramen with concern on the anterior loop is the key factor for increasing the A-P spread and reducing distal cantilever.^11,12  However, complications of surgical trauma to the mental nerve can result in neurosensory disturbances and altered sensation of the lower lip and chin after implant placement.¹³ 

Clinically, the anterior loop cannot be seen but can be detected in panoramic radiographs, cone beam computerized tomography (CBCT), spiral CT, and magnetic resonance imaging. However, in the study by Arzouman et al,¹⁴  the authors concluded that significantly fewer loops were detected in panoramic radiographs as compared with anatomic assessment. Also, significantly shorter anterior loops were identified in panoramic radiographs than when direct measurements were made. Moreover, Kuzmanovic et al¹⁰  showed that 50% of the radiographically observed anterior loops of the mental canal were misinterpreted by observers with panoramic radiography, and 62% of the anatomically identified loops were not observed radiographically. They concluded that panoramic radiographs are unreliable and have high incidences of false-positives and false-negatives in identifying the anterior loop.

Spiral CT and CBCT have also been used for measuring anterior loops in the literature and tend to be more reliable. Kaya et al¹⁵  showed spiral CT scans demonstrated a higher prevalence of mental loops than panoramic radiographs. Uchida et al¹⁶  compared the CBCT and anatomic measurements and concluded that the average length of the anterior loop was 2.2 ± 0.8 mm using CBCT, and no significant differences were found between CBCT and anatomic measurements.

Ngeow et al¹⁷  investigated the visualization of the anterior loop in “different age-groups and genders” using panoramic radiographs, but this study did not facilitate CBCT scans. Different methods of measuring the anterior loop length (AnLL) on CBCT scans were described in the literature. In one study, a software program was used to draw tracing lines of the mental nerve to measure the length of the loop.¹⁸  Some studies measured by counting the first and last cross-sectional 0.5-mm-thick slices after identifying the anterior loop and mental foramen. However, Kaya et al¹⁵  stated that assessing a tomographic cross-sectional cut for the location of the mandibular or mental canals was difficult. In the present study, we presented a simpler method using i-CATVision software to locate the mental foramen and anterior loop on a single oblique transverse view and measure the length of the anterior loop.

The objective of this study is to measure and compare the AnLL of the mental nerve between gender and age groups and to compare the difference between the right and left sides using CBCT. The null hypothesis is that according to the analysis of CBCT scans, age, gender, and right versus left sides have no influence on the anterior loop length.

To estimate the proper sample size for evaluating the difference between groups, StatSoft STATISTICA (power analysis software) with an α of .05 and 2-tailed t test of significance was used. An estimation was made that 176 participants in each gender group and a total of 352 participants would provide 80% power for detecting a small-to-medium effect size (r = .3) in the relationship between AnLL and gender. After setting α = .05, root mean square standardized effect = 0.2, the 1-way analysis of variance (ANOVA) test of significance was used to make a prior estimate that 122 participants in each age group (3 groups) and a total of 366 participants would provide 80% power in the relationship between AnLL and age groups. Combining the results above, 61 women and 61 men in each age group, with a total of 366 participants, were used in the study.

Sixty-one female and 61 male patients were randomly selected for each age group (21–40, 41–60, and 61–80 years) from existing Loma Linda University CBCT scans. All CBCT scans were taken by I-CAT Cone-Beam 3D Dental Imaging System Model (Imaging Sciences International, Hatfield, Penn) under 6- or 13-cm field of view, 20 seconds scan time, and 0.4-mm voxel size. Both right- and left-side AnLLs were measured in each subject. I-CATVision software (Croall Radiography Inc, version 1.8.1.10) was used to measure the AnLL by using multiplanar reconstruction. Multiplanar reconstruction allows images to be created from the original axial plane in either the coronal, sagittal, or oblique plane.

First, the axial cut in the sagittal plane was adjusted to identify the best view of the mental foramen in the axial plane (Figure 1).

Second, the axial plane was rotated until the sagittal cut was parallel to the buccal plate in the area of the mental foramen. Meanwhile, the coronal cut in the axial plane was adjusted to identify the best view of the mental foramen in the coronal plane (Figure 2).

Third, the coronal plane was enlarged, and an oblique cut that passed through the center of the mental foramen was made to create an oblique plane, which allowed visualization of the mental foramen and anterior loop at the same time (Figures 3 and 4). Fourth, this view was enlarged so that measurements could be made to the nearest 0.4 mm (Figure 4). Fifth, a line (line 1) was drawn parallel to the buccal plate, and another line (line 2) was made perpendicular to line 1, which passed through the most anterior point of the anterior loop. (It could be the origin of the incisive canal or the most anterior point of the mental loop curvature.) AnLL was measured from the most anterior point of the mental foramen to line 2 (Figure 4). When the anterior loop was not present or the origin of the incisive canal was located posterior to the mental foramen, negative values were recorded.

Two examiners were involved in the study. Each examiner measured AnLL on both sides of the patient 3 different times. Intra- and Interexaminer reliability were tested by using SPSS intraclass correlation coefficient (Cronbach's alpha value >.8 means optimal agreement, >.9 means excellent agreement). The prevalence of the anterior loop (when AnLL > 0 mm) was calculated. After calculating the mean and standard deviation, the independent t test was used to compare AnLL between gender groups. A paired t test was used to compare AnLL between the right and left sides. One-way ANOVA was used to compare AnLL among the 3 age groups. The α level of all the statistic tests was set as .05.

All measurements were collected for intra- and interexaminer reliability tests. A Cronbach's alpha value of .954 and .976 was obtained in the first and second examiner, showing an excellent agreement. An intraclass correlation coefficient of .904 was obtained comparing 2 examiners, showing excellent agreements as well.

The measurements of the AnLL comparing the right and left side in different gender and age groups are listed in Tables 1–4. The anterior loop was present (length >0 mm) in 624 of 732 hemimandibles (prevalence = 85.2%). The AnLL of the 366 subjects (732 hemimandibles) was 1.46 ± 1.25 mm. The maximal and minimal AnLL were 6.67 mm and −2.87 mm (Figures 5 and 6). The mean AnLL on the right side was 1.47 ± 1.39 mm and on the left side was 1.44 ± 1.39 mm. There was no statistical significance between groups. The mean AnLL was 1.51 ± 1.24 mm for men and 1.40 ± 1.26 mm for women, and no statistically significant differences were found.

Among the age groups, 1-way ANOVA statistical analysis showed a significant difference between groups (P = .000). Both post hoc tests, Bonferroni and Scheffé test, showed the AnLL of the 21–40 year group was significantly different from groups 41–60 and 61–80 years. The mean AnLL in the 21–40 year group (1.89 ± 1.35 mm) was larger than the AnLL in the 41–60 year group (1.35 ± 1.19 mm) and 61–80 year group (1.13 ± 1.08 mm).

In the literature, the average measurements of the AnLL varied. The mean AnLL varied from 0.1 mm¹⁹ to 6.92 mm.²⁰  The reason that the results varied might be due to different diagnostic tools (cadavers, dry skulls, 2D radiographs, and 3D radiographs) different population, (different age, gender, race, dental status), different methods of measuring (direct measurements using probe, calipers), or different methods of interpreting the radiographs. In our study, the mean AnLL of the 366 subjects on CBCT scans (732 hemimandibles) was 1.46 ± 1.25 mm. This corresponded with most of the cadaver studies that used direct measurements. Solar et al²¹  found that the mean AnLL was 1 mm and ranged from 0.5 to 5 mm, Kuzmanovic et al¹⁰  found the mean AnLL was 1.2 mm, and Uchida et al²²  found it was 1.5 mm. All of these were cadaver studies.

Studies by Arzouman et al,¹⁴  Kaya et al,¹⁵  and Misch and Crawford²³  found mean AnLLs of 3.45 mm, 3.75 mm, and 5 mm, respectively. A possible reason for these average AnLLs compared with those of the current study could be the fact that the previous investigators used 2D radiographs as the measuring tools. Kuzmanovic et al¹⁰  concluded that panoramic radiographs are unreliable and have high incidences of false-positives and false-negatives in identifying the anterior loop. They found that radiographic length of the anterior loop of the mental canal can be measured only in radiographs in which the entire course of the mental canal is visualized, from the mandibular canal through the mental foramen (type I, or continuous type; Yosue and Brooks classification²⁴). We applied this concept on the 3D CBCT scans. We think that the most accurate way to measure the anterior loop is when the mandibular canal, the anterior loop, the incisive canal, and the mental foramen can be continuously visualized in the same view.

Cone beam computed tomography was introduced in the early 2000s with its great decrease in dose compared with the conventional CT.^25–27  Poeschl et al²⁸  found that CBCT was as accurate as conventional multislice CT with regard to its use in image-guided implant surgery. Al-Ekrish and Ekram²⁹  stated that the mean of the CBCT absolute errors was even smaller than that of the multi-detector CT absolute errors for the overall data, as well as for the site-specific data. Moreover, Santana et al³⁰  and Uchida et al¹⁶  stated there was no statistically significant difference on the AnLL between anatomic measurements from cadavers and measurements obtained from CBCT images. Although CBCT had been proven to be accurate, studies using CBCT to analyze the average length of the anterior loop still showed contrasting results from 0.89 mm³¹ to 6.92 mm.²⁰  We found that the methods of measuring the AnLL on CBCT were different. Rosa et al³²  determined the mental loop length on panoramic views from CBCT scans, which made it similar to studies using panoramic radiographs. Chen et al²⁰  mentioned that they reconstructed part of the panoramic view along the inferior alveolar nerve canal, but no reference point of measuring the loop was found. Apostolakis and Brown³¹  counted the 0.3-mm vertical cross-section slices to evaluate the length of the loop, and we found it difficult to identify the most anterior part of the mental nerve on the cross-section views, especially when the bone density was poor.¹⁵  The method we used in our study was to locate the mental foramen and anterior loop on a single oblique transverse view. We believed that once the mandibular canal, the anterior loop, the incisive canal, and the mental foramen could be simultaneously visualized in the same view, the measurement would be more accurate. The presence of the anterior loop of mental nerve was defined as any part of the mental nerve located mesial (or anterior) to the mental foramen. In the present study, 85.2% prevalence of the loop was noted. This result was similar to the study of Neiva et al,³³  who found 88% of anterior loop in skulls using a probe, and Kieser et al,³⁴  who found 84% in cadavers. The reason that some studies showed a low prevalence (7% to 55%)^10,32,35,36  might be the difficulty of clearly identifying the most anterior portion of the mental nerve. One disadvantage in the present study is that a medium-resolution CBCT with a scan time of 20 seconds and 0.4-mm voxel size was used. Higher-resolution scans with a scan time of 40 seconds and 0.25-mm voxel size might lead to a clearer image and decrease the errors of measurements. However, both intra- and interexaminer reliability were analyzed in our study, and the statistic result showed an excellent agreement. This result confirmed that the oblique transverse method was a simple and reliable way to measure the anterior loop on a CBCT scan.

Studies that have analyzed the anterior mental loop in cadavers tend to consist of populations with greater age and small sample sizes. Power analysis guided us to collect at least 61 patients in each group, with a total of 366 patients. To the best of our knowledge, this study consisted of the largest sample size of all studies to report on this topic. In the present study, equal subject numbers of different genders with a wide age range from 21 to 80 years were included. The statistical analysis showed that there was no difference between the right and left side or between genders. Uchida et al^16,22  showed no difference between sides, but a larger AnLL was found in men. In our study, we had 5 times more subjects than the previous study, and men showed a mean AnLL of 1.51 mm, which is slightly larger than that of women (1.40 mm) but statistically not significant. Moreover, we found that the AnLL of the 21–40 year age group was significant larger than the 41–60 and 61–80 year groups. This is also consistent with Uchida et al²²  and Ngeow et al,¹⁷  who indicated that the frequency of the anterior loops decreases with age. In 1986, Gershenson et al³⁷  described that with loss of teeth and bone resorption, the mental foramen moves upward closer to the alveolar border. The mental nerve emerged from the mental foramen closer to or at the alveolar border, according to the degree of resorption. Therefore, when teeth are lost, bone resorption occurs, the mental foramen gets closer to the anterior part of the mental loop, and the AnLL decreases. Meanwhile, we found 3.28% of posterior edentulous in the age 21–40 year group, 20.08% in the 41–60 year group, and 27.87% in the 61–80 year group (Table 5). The mean AnLLs of the dentate and edentulous side were 1.55 mm and 1.00 mm, respectively. Therefore, we believe that when the age increases, the number of posteriorly edentulous mandibles increases, the bone resorption increases, and the AnLL and prevalence decrease.

Complications of surgical trauma or pressure on the mental nerve can result in neurosensory disturbances and altered sensation of the lower lip and chin after implant placement.^13,38,39  Ellies and Hawker⁴⁰  reported a retrospective analysis of a multicenter study. They found altered sensation of the mandibular alveolar nerve in 36% of patients after 2 weeks and 13% after 1 year. It has been suggested in the literature that the most distal aspect of an implant should be at least 1 mm,¹⁹  4 mm,¹⁰  or 6 mm²¹ anterior to the foramen to avoid injuring the inferior alveolar and/or mental nerves during an osteotomy based on findings of the prevalence and extent of the anterior loop of the mental nerve. In the present study, the largest AnLL was 6.67 mm. Moreover, the longest loop in the literature was reported by Neiva et al,³³  being 11 mm, followed by Uchida et al¹⁶  with a length of 9 mm. For these cases, following the 1-, 4-, or 6-mm rule would lead us to injure the nerve when placing implants in close proximity to the mental foramina.

On the other hand, if we place every implant 6 mm anterior to the mental foramen, we limit the interforaminal space available for implant placement. The range of the AnLL found in our study is −2.87 mm to 6.67 mm. No studies in the literature have calculated the negative value when the anterior loop was not present. However, Kieser et al³⁴  in 2002 described 5 patterns of emergence of the human mental nerve. Type 1, referred to as a posterior directed emergence, type 2 an anterior directed emergence, type 3 a right-angled pattern of emergence, type 4 to multiple foramina, and type 5 to the alveolus had resorbed, and no evaluation was possible. It is also important to recognize type 2 of mental nerve (for which the anterior loop has a negative value) because the surgeon can place the implant closer to the mental foramen and even posterior to the mental foramen. In the treatment of mandibular arch involving the interforaminal area, a full arch implant-supported restoration involving 4 to 5 implants is a good viable option. To place the implant closest to the mental foramen with concern on the anterior loop is the key factor for increasing the A-P spread and reducing distal cantilever.^11,12  In addition, placing implants in mandibular partially dentate patients involving premolar and molar areas, the mental neurovascular bundle is usually the critical reference point during treatment planning. Therefore, the precise location of the mental loop must be identified using proper radiographic techniques before surgery. We recommended using the oblique transverse reconstruction modality on CBCT to accurately locate and measure the AnLL prior to the implant surgery.

After evaluating 366 CBCT scans of patients:

• 1

  Of the scans, 85.2% revealed the presence of an anterior loop of the mental nerve.

• 2

  A wide range of measurements existed for the AnLL, ranging from −2.87 mm to 6.67 mm, with a mean of 1.46 mm.

• 3

  No statistically significant differences were found between the left- and right-side measurements or between genders, but the 21-40 year age group had a significantly larger AnLL than ages 41–60 and 61–80 years.

When placing implants in close proximity to the mental foramina, caution is recommended to avoid injury to the mental nerve and its neurovascular bundle. The anterior loop of the mental nerve becomes a critical surgical reference point during treatment planning. Because of the wide range of the AnLL observed in our study, no fixed distance mesially or anteriorly from the mental foramen should be considered to be a “safe” distance without the use of 3D imaging. Analyzing CBCT scans using the method described in this article can be a useful tool for avoiding implant surgical complications.

Abbreviations

2D

2-dimensional

3D

3-dimensional

AnLL

anterior loop length

ANOVA

analysis of variance

A-P spread

anterior-posterior spread

CBCT

cone-beam computerized tomography

CT

computerized tomography

MDCT

multi-detector computed tomography

RMSSE

root mean square standardized effect

The authors would like to thank Ashish Sharma, the residents, and faculty members at Advanced Education in Implant Dentistry, Loma Linda University School of Dentistry, for their involvement in the study and Ching-Ho Leu and Udochukwu Oyoyo for their assistance with the statistical analysis.