The AL is a bony prominence on the lateral surface of the mandibular ramus and has been used as a landmark during orthognathic surgery for decades [2, 20, 21]. However, the presence of the AL has been shown to vary across studies, even within those focused on Asian populations, ranging from 32.02 – 100% as demonstrated in Table 5. The methods of study probably impacted the identification of the AL, as it tends to be less visible in CT images compared to dry mandibles. In this present study, the AL could be detected in most samples and the presence of the AL was not shown to be associated with the presence of the LG. Our findings were consistent with the study of Apinhasmit [22] which was performed in Thai mandibles, albeit higher occurrence, and in agreement with Park et al. (Korean mandibles) [23] and Hsiao (Taiwanese patients) [24].

In orthognathic surgery, especially IVRO and inverted-L osteotomy (ILO) techniques, locating the LG from the buccal side of the mandibular ramus is a challenging task. Therefore, it is of importance that the AL, the anatomical landmark indicating the LG position on the lateral side, should be clearly visible and reliable. The accuracy in identification of the AL was assessed by comparing the positions identified by each dental student and the expert. The reliability was also evaluated by comparing the position identified by 3 students. Our assessments showed 86.67 – 90.00% accuracy and 86.67% reliability in identification of the AL which are higher than the studies of Pogrel [15], Yates [16] and Kapur [25] (Table 5). The high accuracy and reliability in our study supports the practicality of this landmark since it could be detected by both novices and experts without difficulty. Nevertheless, owing to the differences in reliability and accuracy among different ethnic groups, multi-center or multi-ethnicity studies should be warranted to confirm our findings before generalizing to other populations.

Although the AL on the lateral surface of the mandible has been commonly used to approximate the LG on the medial surface, several lines of evidence have proven that the AL was not exactly located on the same position as the LG [15, 22, 25, 26]. Indeed, the recent study indicated no correlation between these structures [24]. Our study showed that only 2.5% of the LG exactly corresponded to the AL and 68.33% distributed within 5 mm radius while 86.67% were found within 6 mm radius from the AL. However, while the common position of the LG was shown to be posterior and inferior to the AL [15, 22, 26, 27], the majority of the LG in this present study were located inferior to the AL with greater propensity in the anterior region, giving the percentage of 31.67%. The distribution in this antero-inferior region is comparable to the study of Apinhasmit (27.00%) [22]. Considering the distribution from the AL (Table 6), Pogrel [15] and Apinhasmit [22] showed 43.3% and 84.50%, respectively, of the LG were located within 5 mm radius. The latter is consistent with the distribution within 6 mm in our study. Because the difference of 1 mm is unlikely to be significant in clinical practice, it can be assumed that the LG are mostly located inferior to the AL and within 5 – 6 mm radius in Thai mandibles.

Our study demonstrated that the LG was not always located on the same position as the AL on the lateral side. Moreover, the morphology of the LG has been shown to be diverse among genders and racial groups [28, 29]. Tuli et al. [30] classified such variations into 4 types: truncated, triangular, nodular and assimilated. Due to the diversity of its morphology, using the AL may not be the most suitable approach to identify the LG. Therefore, we also considered the relationship between the posterior border of the mandibular foramen which is another anatomical structure relating to the IANB and the AL. Our result is consistent with the previous studies [16, 23, 31] that the mandibular foramen was usually located posterior and inferior to the AL. Regarding its distribution, Yates et al. found 37.10% of the foramen was within 5 mm radius [16] from the AL whereas only 6.67% were detected in our study. This discrepancy might be because the foramen in the study of Yates et al. referred to the deepest point of the foramen. It is important that the size of the foramen should be taken into account as the width, which was shown to vary from 3 to 11 mm, is used to determine of the length of the horizontal osteotomy in ILO and BSSRO [32]. Consequently, the anterior border of the foramen was also investigated. Our result indicated that the AMF did not represent the position of the AL which is in agreement with the previous report by Hosapatna et al. [14].

We demonstrated that the mandibular foramen did not correspond to the AL especially its posterior border which was located further than other reference points from the AL. Therefore, the speculation that the IANB compresses the medial surface of the ramus and leads to a prominence on the lateral surface when the bundle passes through the foramen may not hold true. Reitzel et al. [18] and Hogan and Ellis [17] proposed another hypothesis that the AL is a bony elevation occurring in response to the tendon of the deep head of masseter muscle which inserts at the midpoint of the ramus. The elevation was called the masseteric apical bump [18, 19]. In addition, the bony ridge similar to that usually found at the insertion of a muscle was clearly observed in the mandible of primates and canines at the same area as the masseteric apical bump in human mandible [17]. This assumption is thus more likely to underlie the cause of the AL.

Apart from the horizontal plane, we also investigated the location of the reference points in relation to the reference plane which was parallel to the occlusal plane and passed through the coronoid notch, or the plane that serves as a guide for inferior alveolar nerve block. This reference plane is assumed to be at the level of the mandibular foramen and lies approximately 6–10 mm superior to the occlusal plane [33]. Our finding showed that the AL and the PMF were inferior to the reference plane with the distances of 2.80 ± 3.69 mm and 8.36 ± 4.45 mm, respectively. Since our results also indicated that the LG was 1.58 ± 3.26 mm inferior to the AL, we could estimate that the LG was positioned at 4.5 mm inferior to the reference plane, or approximately 5.5 mm superior to the occlusal plane. Consistent with our data, a previous study on Thai mandibles reported that 78.52% of the LG was located 4.5 mm superior to the occlusal plane [29]. To our knowledge, our study is the first to report the AL and the PMF relative to the plane used in the inferior alveolar nerve block, a plane well recognized by most dental surgeons. These relationships can be applied to estimate the position of the irregular shaped AL and the posterior border of the mandibular foramen when the AL cannot be identified.

The recommended distance to avoid damage to the IANB in IVRO is 7 – 8 mm anterior to the posterior border of the ramus [9] or 8 mm posterior to the AL [25]. In addition, Werther and Hall [34] proposed that the length of the ramus posterior to the osteotomy line should be at least 6 mm to preserve the viability of the remaining bone. When the AL is used as a reference point, our results suggested that in Thai mandibles, the distance of 10 – 11 mm posterior to the AL is considered safe for the vertical osteotomy. This distance is far enough to avoid IANB while the bone viability can still be preserved. Due to the variations in the size of the ramus, the ratios of the measurements were also calculated. Our PMFw/W ratio is also consistent with Park et al. and Chen et al. [23, 35] indicating that the region suitable for IVRO is the point at the posterior one third of the ramus, or at 30% of the horizontal length of the ramus from the posterior border. However, the location of the AL and the mandibular foramen can vary among different facial morphologies i.e. short and broad face, or long and narrow face, and skeletal patterns i.e. skeletal class I, II and III [10, 36]. Therefore, careful considerations should be taken when using the AL as a sole reference point of the mandibular foramen. The estimation of the mandibular foramen in relation to the AL using a cone beam computed tomography (CBCT) can also endorse the use of the AL in clinical setting [27, 35].

The limitation of this study was that the mandibles were of unknown age and sex which are the factors affecting the morphology of the mandibular ramus and the structures on the ramus. A recent study showed that the location of the AL and the dimensions of the ramus were correlated with sex [10], probably due to the smaller size of female mandibles compared to male mandibles. Another study also indicated the tendency for the mandibular foramen to shift superiorly with advancing age [37]. Nevertheless, whether the location of the AL changes in a similar pattern with the foramen requires further clarifications. In clinical practice, the ages of patients undergoing orthognathic surgery are generally younger than those of the dried mandible in our study. This age difference should be considered when estimating the location of the mandibular foramen.