Mandibular surgery for the treatment of dentofacial alterations began in the 19th century, with studies by Hullihen [8] in 1849, changing over time, with more important advances after the technical modifications by Obwegeser [9], Dal Pont [18] and later Hunsuck [12] and Epker [4], when it began to be performed intraorally and with greater bone contact, increasing its predictability and reliability, especially when associated with rigid osteosynthesis [2, 3, 4, 12, 22, 23, 24, 25, 26, 27, 28, 29].
The bilateral mandibular sagittal osteotomy technique provides adequate bone contact and good results, but it still has some drawbacks, especially related to paresthesia in the inferior alveolar and sometimes lingual nerves, to fractures that are sometimes improper and ascending to the mandibular condyle and TMJ, bone interference when mandibular positioning can cause instabilities or even recurrences, with osteosynthesis failure [1, 3, 4, 5, 7, 9, 12, 18, 20, 25].
Over time, even for such a highly regarded technique, modifications that could lead to improvements and solve these problems are researched. It was correlated to why these complications occurred, their frequency, and ways to predict to avoid them were sought. Many studies modified the extent of this surgical tracing, the way of positioning its fractured stumps, the mandibular bone thickness in its anatomical segments, and how this could weaken or predispose to possible trans and post-surgical complications [1, 2, 5, 9, 10, 17, 18, 21, 23, 28].
From a technical point of view, obtaining the sagittal and vertical fracture line from the buccal side of the mandible is relatively easy to perform, since its visualization is well observed. The lingual fracture line, after cutting with a saw, drill, or ultrasonic tip, although it can be visualized in its medial design above the lingula, as long as with adequate anatomical distance, has a limitation regarding the complement of the fracture. Thus, despite being delimited and drawn with a drill horizontally, when proceeding with the mandibular surgical fracture itself or bone separation, it is not possible to perfectly visualize how this trace occurs on the lingual table of the ramus and mandibular body, due to the presence of anatomical structures [1, 3, 5, 7, 12, 18, 20, 22, 25, 26].
It is not known for sure how these variables of tracings or lingual intercurrences can interfere with the results of sagittal mandibular orthognathic surgery. For years, radiographs were the only way to show the result of post-surgical movement. There was, however, the impossibility of fully verifying its entire course, especially lingually, and how this surgical feature had behaved [1, 12, 14, 16, 18, 20, 21, 23].
The advent of computerized tomography made it possible to advance both in the planning of surgeries and the verification of their results. It made it possible for the anatomical tracings obtained through surgical techniques to be checked, measured along the way, visualized if there were alterations, and even published [1, 5, 7, 8, 17, 23, 24, 26].
The modifications in the techniques continued to occur because they are part of the evolution of science, but now they could be properly analyzed if they really were what they proposed and if all their results and modifications happened in their entirety. The techniques of Hunsuck [12], Dal Pont [18], and Obwegeser [9] were studied in many studies correlating with the path of the inferior alveolar nerve, with the mandibular bone thickness, with the use of drills, saws, or ultrasonic tips, with the design and pattern of bone separation after osteotomy and how they behaved, visualized through computerized tomography [22, 26, 27, 28]
In addition to the technical variable, Aarabi et al. [20 ]and Bell and Schendel [2] were authors who correlated the original osteotomy design and areas of bone fragility, increases in thickness in each segment, or simply a certain tendency to fracture, which could interfere with the original designs. This should be taken into account when performing the surgery and especially in the post-surgical period [2, 3, 4, 7, 8].
Arnett [1] was the first to idealize that shorter medial or lingual osteotomies, with less tissue detachment, less lingual detachment, and, consequently, less bone interference could lead to better results, mainly sensory and with fewer risks of undue fractures and ascending fragmentation for condyle and TMJ [2, 5, 6].
The lingual short split mandibular sagittal osteotomy, as described by Santana and Souza [5], follows a trend towards the sagittal osteotomy of the ramus. It presents an important modification with a medial/lingual cut, according to the concepts of Arnett [1], with small technical modifications, being parallel to the mandibular plane and below the lingula height, performed with a drill and surgical saw in its lingual, sagittal and vertical cuts. The shorter cut, pre lingula, allows for less bone interference, less detachment of the pterygoid-masseteric belt, and less possibility of undue fractures ascending to the condyle. Clinically, it was also reported by the authors to cause a shorter duration of inferior alveolar nerve paresthesia, due to less detachment and lingual manipulation [7, 14, 23].
Carrying out the tomographic study of this technique, based on the aforementioned evolution, would be a natural way to verify if the surgical route proposed in the description and its advantages happened. Muto et al. [15] and Plooji et al. [14] carried out innovative studies, specifically to determine the lingual fracture tracing, already emphasizing the possibilities of variation that occurred for the Hunsuck technique [12], analyzing the behavior, and establishing classifications. These classifications showed that although the Hunsuck technique [12] is widely used by surgeons in general and widely published as a predictable technique, it has shown to be variable in its lingual cut, which could somehow be influencing the results of postoperative symptoms, especially sensitivity.
In this paper, we carried out a bilateral tomographic study of the operated mandible of 30 patients, totaling 60 sagittal mandibular lingual short split osteotomies, according to the description and surgical protocol by Santana e Souza [5]. The study was carried out following the work of Muto et al. [15] and Plooij et al. [14], as well as their and other authors methodology of tomographic studies.
It was design a the classification from 1 to 4 based on the design of the surgical tracing of the osteotomy, being type 1 when it was exactly like the original technical proposal, type 2 with little variation in tracing, and type 3 with no fracture to the lower base of the mandible or with it following from the posterior cortex of the ramus and type 4 is similar in design to type 2, but with fragmentation. Clinically, these tracing differences were not visualized by the surgeons, which did not cause any interference or change in conduct during the surgery. However, this study was prospective based on computerized tomography and the visualization of tracing alterations was the objective and focus of the work. There was no need, in the time determined for the collection and analysis of the exams, to change the behavior of the patients, since they had an excellent postoperative period.
Plooji et al. [14] published 4 types of tracing variations for Hunsuck's osteotomy [12]. These variables occurred on the lingual region of the mandibular ramus and were compatible with the original tracing, according to the technical description by Hunsuck [12] in 51.25% of the CT scans examined. The study by Muto et al. [15] also showed that there would be tracing variables, which demonstrated that the Hunsuck technique [12] was not 100% uniform. For Muto et al. [15], the more anterior and shorter the medial cut, the smaller variations would be found.
As Plooji et al. [14], we found 4 classifications, but in no case of the lingual short split osteotomy tracing was their ascendance to the condyle and TMJ, as in his variable LSS4, corresponding to 2.5% of his sample. The variable 4 classification of the lingual short split (with 1 osteotomy or 1.66% of our sample) was a small fragmentation of the lower part of the descending branch, which in no way compromised the result of the surgery or its technical progress.
We found similarity with the original technical description (type 1 classification) in 73.33% of the drawings, a percentage well above that described by Plooji et al. [14] for the original Hunsuck technique [12]. Therefore, the predictability is much higher for the lingual short split technique and the possibility of alterations is lower. The type 1 classification design of the lingual short split osteotomy was transcanal, similar to the LSS3 design by Plooij et al. [14], but with onset before the lingula and below its level.
The type 2 classification of our study was viewed 11 times or 18.33% of the sample. This tracing is similar to the original technique drawing, but with a descending trace a little further towards the back. There is a lot of similarity to the LSS1 classification by Plooij et al. [14], but again the design of the lingual short split osteotomy occurs anteroinferior at the level of the lingula and below the mandibular canal, without bordering it. It is a very stable design and very similar to Santana e Souza's technical proposal [5].
Type 3 classification was found in 4 osteotomies or 6.66% of the sample. This tracing had a very similar design to that of Dal Pont 10 and with the LSS2 classification by Plooji et al. [14]. It can be considered an important variant of the original technical design, since the descending tracing occurred from the posterior border of the mandible in an anteroinferior course, without loss of result.
It is important to point out that eventual variations in the bone thickness of the ramus, amount of cortical or medullary bone, and volumetric alterations of the mandible were not the object of this study, as well as whether the osteotomy was for correction of Angle Class I, II or III patients was not considered in the results. Therefore, any mandibular anatomical alterations evaluated in similar studies such as those by Aarabi et al. 1, Bell and Schendel 3, and Cunha et al. 9 were not the objective of this work.
Regarding gender, there was a slight predominance in the sample of the classification of the tracing type 1 for females, type 2 for males, type 3 for the same number, and 1 type 4 osteotomy for males. Perhaps this trend towards the classification of type 1 in females is due to the greater amount of medullary bone and less cortical bone, but we emphasize that it was not the focus of the study in this work and, therefore, this small prevalence should be better studied in the future.
Regarding laterality, right or left osteotomy, we observed that the type 1 classification was observed in 22 right and 22 left osteotomies or 73.33% of the total sample. We found a small prevalence of type 2 tracing classification on the left side and a significant increase in the percentage of type 3 classification on the right side. Type 4 occurred only once on the left side. It was not possible to correlate whether the side factor was a determinant for the occurrence of variables in the drawings, except for type 3, which had a slightly higher incidence. The occurrence of type 4 classification with small fragmentation, which occurred on the left side in only 1 case, may have been an occasional or accidental event.
We also observed in the CT scans that types 1 and 2 have very similar designs and trace follow-up with little variation. As type 1 with 73.33% and type 2 with 18.33% totaled 91.66% of the sample, we can idealize that the lingual short split technique complies with the criteria of eligibility and predictability, with a high percentage of reliability concerning its original design or minimal variation, without interfering with its results in terms of stability or loss of osteosynthesis. According to what Santana et al. [5] published, the tracing follows the lines of mandibular fragility, and this can be seen in the tomographic analysis, where there was a high percentage of results compatible with the original design.
We also performed the chi-square analysis, to observe whether the drawings found had a statistically significant difference. We found a small margin of variation in relation to gender, laterality, and age, but this analysis concluded that considering the number of total osteotomies, 60, only a p lower than 0.05 would be significant. Thus, it can be concluded that, in these criteria, there were no statistical differences that could be important in the study of surgical tracing variation.