Journal of the Formosan Medical Association (2016) 115, 981e990

Available online at www.sciencedirect.com

ScienceDirect

journal homepage: www.jfma-onl ine.com

ORIGINAL ARTICLE

Anatomical position of the mandibular canalin relation to the buccal cortical bone inChinese patients with different dentofacialrelationships

Chun-Yuan Huang a,b, Yu-Fang Liao c,d,*

a Division of Plastic Surgery, Department of Surgery, Wan Fang Hospital, Taipei Medical University,Taipei, Taiwanb Department of Surgery, School of Medicine, College of Medicine, Taipei Medical University, Taipei,Taiwanc Department of Craniofacial Orthodontics, Chang Gung Memorial Hospital, Taoyuan, Taipei, Taiwand Graduate Institute of Dental and Craniofacial Science, College of Medicine, Chang Gung University,Taoyuan, Taiwan

Received 30 June 2015; received in revised form 25 September 2015; accepted 5 October 2015

KEYWORDSinferior alveolarnerve;

mandibular canal;prognathism;retrognathism;sagittal split ramusosteotomy

Conflicts of interest: All contributi* Corresponding author. Department

Taoyuan County, Taipei, Taiwan (R.O.C33302, Taiwan (R.O.C.).

E-mail address: yfliao0125@yahoo.

http://dx.doi.org/10.1016/j.jfma.2010929-6646/Copyright ª 2015, FormosCC BY-NC-ND license (http://creative

Background/purpose: The purpose of this study was to determine the position of the mandib-ular canal in relation to the buccal cortical bone in Chinese patients with three dentofacial re-lationships: normal dentition, retrognathism, and prognathism.Methods: Cone-beam computed tomography and lateral cephalograms of patients with normaldentation, retrognathism, and prognathism (nZ 32 each group) were reviewed. Measurementsof the shortest distance from the outer/buccal edge of the mandibular canal to the inner sur-face of the buccal cortex, and the distance from the lingula of the ramus to the dorsal root ofthe first molar were recorded.Results: No significant difference was observed between the three groups in the distribution ofcontact or fusion of the mandibular canal, or in the course of the mandibular canal on the rightor left side. When the shortest distance at the lingula on the left side was >2.1 mm, no in-stances of contact or fusion were observed. On the right side, 100% of the patients had no con-tact or fusion when the shortest distance was >2.7 mm at the lingula.

ng authors declare no conflicts of interest.of Craniofacial Orthodontics, Chang Gung Memorial Hospital, No.123, Dinghu Rd., Guishan Township,.); College of Medicine, Chang Gung University, No.259, Wenhua 1st Rd., Guishan Dist., Taoyuan City

co.uk (Y.-F. Liao).

5.10.004an Medical Association. Published by Elsevier Taiwan LLC. This is an open access article under thecommons.org/licenses/by-nc-nd/4.0/).

982 C.-Y. Huang, Y.-F. Liao

Conclusion: The shortest distance from the outer/buccal edge of the mandibular canal to theinner surface of the buccal cortex measured at the lingula can predict contact or fusion. Duringsagittal split ramus osteotomy, great care should be observed at the point halfway betweenthe lingula and the anterior ramus border where the inferior alveolar nerve is the closest tothe cortical bone.Copyright ª 2015, Formosan Medical Association. Published by Elsevier Taiwan LLC. This is anopen access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

Introduction

Sagittal split ramus osteotomy (SSRO) is a common operationused to correctmandibular deformities such as prognathism,retrognathism, and asymmetry.1 The procedure can result ingood functional and cosmetic outcomes; however, neuro-sensory disturbance of the inferior alveolar nerve (IAN) iscommonwith an incidence of 9e85%.2,3 Sensory alteration inthe IAN that is confirmed with neurosensory testing afterSSRO ranges postoperatively from 54% to 86% at 4e8 days,41% to 75% at 1month, 33% to 66% at 3months, 17% to 58% at 6months, and 15% to 33% at 1 year.4

Injury to the IAN during SSRO can occur because of nervelaceration during cortical bone dissection, the nerve maybe torn during splitting, or the nerve may be damaged byinterosseous fixation.1,5 Unfavorable fractures during SSROand contact between the mandibular canal and theexternal cortical bone are the primary causes of IANinjury.6,7 Unfavorable fractures occur mostly because offusion of the medial and lateral cortical plates with nocancellous bone in between.6,7 The incidence of unfavor-able fractures is as high as 50%, and complete transectionof the IAN can occur.8e11 Yamamoto et al.12 showed thatthe mandibular canal was in contact with the externalcortical bone on 10 (25%) sides, and that neurosensorydisturbance occurred on all sides with an incidence signif-icantly greater than the 20% incidence of the 30 (75%) sideswithout contact between the canal and external corticalbone.

Anatomical variations of the IAN are therefore importantto the clinician. The mandibular canal courses lingually tothe roots of the second and third molars, adjacent to theroots of the first molar, and laterally to the roots of thepremolars.13 In the area of the mandibular foramen, theIAN occupies nearly the entire cancellous space betweenthe lingual and buccal cortical plates while maintaining aclose relation to the lingual plate and, as it approaches themental foramen the mandibular canal, it turns sharplymedially to laterally towards the foramen.14 Tsuji et al15

detected 16 of 70 skeletal class III rami having contact orfusion of the mandibular canal; in many patients, it extendsfrom the mandibular foramen to the mandibular angle.

Performing computed tomography (CT) before surgery isuseful to obtain the location of the mandibular canal, andto determine the distribution of the mandibular ramuscancellous bone. The use of cone-beam CT (CBCT) in den-tomaxillofacial imaging has several advantages over con-ventional CT such as x-ray beam collimation, reduced

effective radiation dose, and fewer artifacts.16 It also hashigh accuracy and reproducibility.17e20

To date, the course of the IAN canal has not beencompared in the three classes of dentofacial relationships(i.e., normal dentition, retrognathism, and prognathism).Thus, the purpose of this study was to determine theanatomical position of the mandibular canal in relation tothe buccal cortical bone in Chinese patients with the threedifferent classes of dentofacial relationships using CBCTrecords to provide information on the relative distance ofthe IAN to the bone cuts performed during SSRO. This in-formation may help avoid injury to the IAN during SSRO.

Materials and methods

This study was approved by the Institutional Reviewer Boardof our hospital. The requirement for informed patient con-sent was waived because of the retrospective nature of thisstudy. The medical records of patients with Class I, II, and IIIdentofacial relationships who received CBCT scans andlateral cephalograms from 2008 to 2010 were retrospectivelyreviewed. The collection was stopped when the targetnumber of patients was reached (i.e., 16 male patients and16 female patients in each group). Inclusion criteria weremen and women who were aged 18e45 years old. Exclusioncriteria were pathology in the mandible (e.g., cysts, tu-mors), a history or evidence of previous surgery of themandible and cleft lip/palate, and obvious asymmetry of themandible. Patients with no first and second molars were alsoexcluded because this would influence teeth orientation.

Dentofacial skeletal patterns were determined bylateral cephalogram (Figure 1A). Class I (i.e., normalanteroposterior relationship of the mandible to the maxilla)was defined as a point A-nasionepoint B (ANB) angle of0e4�. Class II (i.e., retrognathism; posterior relationship ofthe mandible to the maxilla) was defined as an ANB angle>4�. Class III (i.e., prognathism; anterior relationship of themandible to the maxilla) was defined as an ANB angle <0�.

The CBCT scans were examined using iCAT Vision 1.62software (Imaging Sciences International, Hatfield, PA,USA) with a voxel size of 0.4 mm � 0.4 mm � 0.4 mm. Thesoftware allows multiplanar reconstruction of the axialthree-dimensional (3D) image dataset into appropriateplanes for measurements. Before the measurement, thethree planes were oriented. In the coronal view, thesagittal plane was adjusted so that it bisected the skull.The axial plane was adjusted parallel to the occlusal planeof the mandibular posterior teeth on both sides in the

Figure 1 Measurement of the distance. (A) A-nasionepoint B (ANB) angle measurement. (B) The cone-beam computed tomog-raphy (CBCT) slice thickness is 2 mm (parallel yellow lines). (C) Cross-sectional view of a representative mandible slice, as shown in(B). The red line indicates the shortest distance of the outer/buccal edge of the mandibular canal to the inner surface of the buccalcortex. (D) Line A is the vertical distance from the level of the lingula to the ramus notch. Line B is the vertical distance from thelevel of the lingula to the point of fusion between the medial and lateral cortical plates.

Figure 2 Measurements from the lingula of the ramus to thedorsal root of the first molar. The slice thickness is 2 mm.R Z anterior border of ramus.

Mandibular canal in dentofacial relationships 983

sagittal view, and it was centered at the midpoint of thedistance between the two first molars. The coronal planewas at a right angle to the axial plane in the sagittal view,and was adjusted to bisect the mesial root of the first molartooth in the axial view. The slice thickness of the scans was2 mm (Figure 1B).

The shortest distance was measured from the outer/buccal edge of the mandibular canal to the inner surface ofthe buccal cortex (Figure 1C) and from the lingula of theramus to the dorsal root of the first molar on the left andright sides (Figure 2).7,21 Contact or fusion of the mandib-ular canal was defined, as previously described (Figure 3).15

Measurements were obtained to one decimal place using adigital millimeter scale. The IAN canal courses to the innersurface of the buccal cortex. It was traced every 2 mm fromthe lingula of the ramus (i.e., 0%) to the level of the dorsalroot of the first molar (i.e., 100%). The values were recordat every 10% interval between 0% and 100%. When exactvalues could not be measured, the percentile values wereestimated using the interpolation method. The verticaldistance from the level of the lingula to the ramus notch,

Figure 3 Classification of the mandibular canal. (A) Separatedthe canal does not touch the inner surface of cortical bone. (B)Contact typedcontact is evident between the outer surface of the canal and the inner surface of the cortical bone. (C) Fusiontypedthe outer surface of the canal is not visible.

984 C.-Y. Huang, Y.-F. Liao

and the vertical distance from the level of the lingula to thepoint of fusion between the medial and lateral corticalplates are illustrated in Figure 1D.

Statistical analysis

Continuous data were presented by the mean � standarddeviation, and categorical data were presented by thenumber and percentage. For continuous data, one-wayanalysis of variance with Bonferroni post-hoc tests forpair-wise group comparisons were performed to comparethe three groups. For categorical data, Fisher’s exact testwas performed. Receiver operating characteristics (ROC)analysis was used to determine if contact or fusion of themandibular canal could be predicted by the shortest dis-tance from the outer/buccal edge of the mandibular canalto the inner surface of the buccal cortex measured at thelingula. Two types of optimized cut-off points were deter-mined: (1) the point with a maximum of sensitivity þspecificity e 1; and (2) the point with maximum specificitywhen the sensitivity was equal to 1.0. Statistical analyseswere performed using SPSS 15.0 statistics software (SPSS

Inc., Chicago, IL, USA). A two-tailed value of p < 0.05 wasconsidered statistically significant.

Results

Ninety-six patients were included with 32 individuals (16women and 16 men) in each of the three groups (i.e.,normal dentition, retrognathism, and prognathism).Detailed demographic and measurement data of the threegroups are shown in Table 1. Individuals in the prognathismgroup were significantly younger than those in the retro-gnathism group (mean age, 23.4 years vs. 29.8 years;p < 0.001).

Lingula tip to the ramus notch and fusion of thebuccal and lingual plates

There was no difference in the distance from the lingula tipto the ramus notch between the three groups (for all,p > 0.05). There was no difference between the threegroups in the distance from the lingula tip to fusion of the

Table

1Comparisonsofpatients

withnorm

aldentation,retrogn

athism,andprogn

athism.

Allpatients

(nZ

32)

pFemale

patients

(nZ

16)

pMale

patients

(nZ

16)

p

Norm

al

Retrogn

athism

Progn

athism

Norm

al

Retrogn

athism

Progn

athism

Norm

al

Retrogn

athism

Progn

athism

Age

(y)

26.6

(6.9)

29.8

(8.3)

23.4

(4.2)a

0.00

1b

Lingu

latipto

theramusnotch(m

m)

Right

14.3

(2.6)

14.0

(2.8)

15.5

(3.0)

0.08

213

.1(1.7)

13.7

(3.0)

14.8

(2.7)

0.16

715

.4(2.9)

14.2

(2.7)

16.2

(3.3)

0.19

6Le

ft14

.3(2.3)

13.9

(2.8)

15.3

(2.8)

0.12

113

.9(1.7)

13.2

(3.1)

14.7

(2.5)

0.25

214

.8(2.8)

14.6

(2.5)

15.9

(3.0)

0.40

3Lingu

latipto

thefusionofthebucc

alandlingu

alplates(m

m)

Right

6.7(1.6)

5.9(1.6)

6.1(1.3)

0.07

26.8(1.6)

6.5(2.0)

6.0(1.1)

0.42

66.7(1.7)

5.2(0.8)c

6.1(1.6)

0.01

8b

Left

6.4(1.6)

6.4(1.7)

6.4(1.4)

0.99

46.8(1.5)

6.2(1.4)

6.2(1.4)

0.42

46.0(1.6)

6.6(1.9)

6.6(1.5)

0.51

3Contact

orfusionofthemandibularca

nal

Right

6(18.8%

)8(25.0%

)7(21.9%

)0.95

13(18.8%

)2(12.5%

)3(18.8%

)1.00

03(18.8%

)6(37.5%

)4(25.0%

)0.60

9Le

ft5(15.6%

)7(21.9%

)7(21.9%

)0.85

21(6.3%)

3(18.8%

)2(12.5%

)0.85

94(25.0%

)4(25.0%

)5(31.3%

)1.00

0

Data

are

presentedbythemean(standard

deviation).Howeve

r,sexispresentedbythenumberoffemaleandmale

patients,andco

ntact

orfusionofthemandibularca

nal

ispresentedby

thenumber(percentage

).aTheage

ofpatients

intheprogn

athism

groupwas

sign

ifica

ntlyyo

unge

rthanin

theretrogn

athism

group.

bAva

lueofp<

0.05

indicatesasign

ifica

ntdifference

betw

eenthethreegroupsofpatients.

cThedistance

from

thelingu

latipto

thefusionofthebuccalandlingu

alplatesontherigh

tsideofmenin

theretrogn

athism

groupis

sign

ifica

ntlyshorterthanthatofthenorm

algroup.

Mandibular canal in dentofacial relationships 985

buccal and lingula plates, except for the right side of malepatients in whom the distance from the lingula tip to thefusion of the buccal and lingual plates in the retrognathismgroup was significantly shorter than the distance in thenormal group (mean distance, 5.2 mm vs. 6.7 mm;p Z 0.016). The maximum and minimum distance from thelingula tip to the fusion of the buccal and lingual plates was10.3 mm and 3.6 mm, respectively.

Mandibular canal and the IAN course

There was no significant difference between the threegroups in the distribution of contact or fusion of themandibular canal (p > 0.05; Table 1). The shortest distancefrom the outer/buccal edge of the mandibular canal to theinner surface of the buccal cortex on the right side or leftside was similar between the three groups (Figure 4).

On the left side in the normal anteroposterior group, theaverage shortest distance from the outer/buccal edge ofthe mandibular canal to the inner surface of the buccalcortex was lowest at 20% of the distance of the lingula ofthe ramus to the dorsal root of the first molar, and greatestat 70% of the distance of the lingula of the ramus to thedorsal root of the first molar. The other two groupsexhibited similar trends, but the lowest and highest levelsin the retrognathism group occurred at 10% and 60%,respectively, of the distance of the lingula of the ramus tothe dorsal root of the first molar. The lowest and highestlevel in prognathism group were at 20% and 70%, respec-tively. Most R points (i.e., anterior border of the ramus)were at 40% and 50% (81.3%). Fifteen (15.6%) R points and 3(3.1%) R points occurred at 30% and 60%. In 86.5% (83/96) ofpatients, the R points were located after the points ofminimum shortest distance of the outer/buccal edge of themandibular canal to the inner surface of the buccal cortex.

On the right side in the normal anteroposterior group,the average shortest distance of the outer/buccal edge ofthe mandibular canal to the inner surface of the buccalcortex was lowest at 20% of the distance of the lingula ofthe ramus to the dorsal root of the first molar, and greatestat 80% (Figure 4). The other two groups exhibited similartrends, but the lowest and highest levels of the retro-gnathism group occurred at 10% and 70%, respectively. Thelowest and highest levels of the prognathism groupoccurred at 20% and 70%, respectively. More than one-halfof the R points occurred at 40% (56.3%); 21 (21.8%) Rpoints occurred at 50%; and 17 (17.7%) R points and 4 (4.2%)R points occurred at 30% and 60%, respectively. In 91.7%(88/96) of patients the R points were located after thepoints of the minimum shortest distance.

Contact or fusion of the mandibular canal

Nineteen (19.8%) of the 96 patients had contact or fusion onthe left side (Figure 5). Most (89.5%) of the contact orfusion occurred at 0e20%: seven patients (0%), five patients(10%), and five patients (20%). The remaining two patientshad contact or fusion at 30% and at 20e30%. All R pointsoccurred at 30e50%: four patients (30%), 10 patients (40%),and five patients (50%). All contact or fusion on the left sideoccurred before the R point.

Figure 4 Comparison of the inferior alveolar nerve course in subjects with normal dentation, retrognathism, and prognathism.Each point and error bar represents the average level and standard deviation of 32 subjects in the corresponding group. The graybars represent the number of patients for the distribution of R point. (A) For the 96 patients, 15, 45, 33 and 3 patients had their Rpoints on the left side at 30%, 40%, 50%, and 60%, respectively. (B) For the 96 patients, 17, 54, 21, and 4 patients had their R pointson the right side at 30%, 40%, 50%, and 60%, respectively.

986 C.-Y. Huang, Y.-F. Liao

Twenty-one (21.9%) of the 96 patients had contact orfusion on the right side. The R points were at 30%, 40%, and50% in four patients, 12 patients, and five patients,respectively. Most (85.7%) contact or fusion occurred at0e20%: 12 patients (0%), four patients (10%), and two pa-tients (20%). The other three patients had contact or fusionat 40% (two patients) and 20e30% (one patient). Except fortwo patients in whom contact or fusion and the R pointwere at the same site, all contact or fusion on the right sideoccurred before the R point.

Receiver operating characteristic analysis

Receiver operating characteristic analysis was performed todetect contact or fusion of the mandibular canal by the

shortest distance from the outer/buccal edge of themandibular canal to the inner surface of the buccal cortexmeasured at the lingula. The area under the ROC curves(AUC) was 0.953 (p < 0.001) and 0.977 (p < 0.001) fordetecting contact or fusion of the mandibular on the leftside and right side, respectively (Figure 6).

On the left side, a cut-off point of 1.4 mm obtained amaximum sensitivity and specificity of 0.900 and 0.855,respectively (Figures 6A and 7A). Thus, 90% of patients withcontact or fusion on left side had the shortest distance(i.e., <1.4 mm, which is a positive outcome) and 85.5% ofpatients without contact or fusion on the left side had theshortest distance (i.e., >1.4 mm, which is a negativeoutcome). The negative predict value (NPV) was 97.0% (65/67), which indicated that 65 of 67 patients with the shortestdistance at the lingula (i.e., >1.4 mm) did not have contact

Figure 5 The inferior alveolar nerve course of the patientswith contact or fusion. Each point represents the averagelevel. The gray bars represent the number of patients for thedistribution of R point. (A) Among 19 patients with contact orfusion on the left side, four patients, 10 patients, and fivepatients had R points at 30%, 40%, and 50%, respectively. (B)Among 21 patients with contact or fusion on the right side, fourpatients, 12 patients, and five patients had their R points at30%, 40%, and 50%, respectively.

Mandibular canal in dentofacial relationships 987

or fusion. At the cut-off point of 2.1 mm, the sensitivity andspecificity were 1.000 and 0.658, and the NPV was 100%(50/50), which indicated that all patients with the shortestdistance at the lingula (i.e., >2.1 mm) did not have contactor fusion.

On the right side, a cut-off point of 1.25 mm obtained amaximum sensitivity and specificity of 0.810 and 0.893,respectively (Figures 6B and 7B). Thus, 81% of patients withcontact or fusion on the right side had the shortest distance(i.e., <1.25 mm, a positive outcome), and 89.3% of patientswithout contact or fusion on the right side had the shortestdistance (i.e., >1.25 mm, a negative outcome). The NPVwas 94.4% (67/71), which indicated that 67 of 71 patientswith the shortest distance at the lingula (i.e., >1.25 mm)

did not have contact or fusion. At a cut-off point of 2.7 mm,the sensitivity and specificity were 1.000 and 0.400,respectively, and the NPV was 100% (30/30), which indi-cated that all patients with a shortest distance (i.e.,>2.7 mm) at the lingula did not have contact or fusion.

Discussion

Unfavorable fractures can result in an IAN injury andreportedly occur in up to 50% of SSRO cases; they primarilyresult from fusion of the medial and lateral cortical plateswith no cancellous bone in between.6,7,11 Muto et al.7

report that the safest region in which to establish themedial osteotomy line is just superior and 5e6 mm poste-rior to the lingula, and with the line directed slightly infe-rior. Another investigator reports that a desirable splittingpattern occurs when a short lingual cut just above the lin-gula and a lateral bone cut of the mandibular angle arecreated, and extended to the inside through the inferiorborder of the mandible.22

Based on our results, no significant difference wasobserved in the distance from the lingula tip to the fusion ofbuccal and lingual plates between the three groups, whichindicated a similar chance of unfavorable fractures be-tween the three groups during medial osteotomy. A medialosteotomy line 3 mm superior to the lingula and carried tothe depth of the medial surface of the buccal cortex issuggested because the minimum distance from the lingulatip to fusion of the buccal and lingual plates is 3.6 mm.

Chances of an unfavorable fracture are reduced if themandible is split in a region where it is sufficiently thick.Nagadia et al23 examined the anatomical position of themandibular canal with respect to cortical bone and molarteeth in Chinese individuals and found that the mandibularcanal was the farthest from the buccal cortex at the secondmolar region (mean distance, 6.79 mm; minimum distance,4.80 mm), and the mandibular body was thickest (11.9 mm)in the region of the second molar. Nagadia et al23 concludedthat the vertical buccal cut for SSRO should be near thesecond molar where the bone is thickest and the mandib-ular canal is furthest from the buccal cortex, and that thesafe depth for the vertical buccal cut is 4.8 mm, whichcorresponds to the minimum horizontal distance. In addi-tion, when contact or fusion is present, the IAN is easilyinjured when the bone is separated from the neural tube.7

The position of the IAN canal at the ascending ramus isrelated to neurosensory disturbance after SSRO.11,24 Basedon our observations (Figure 4), IAN was nearest to thecortical bone at 10e20% of the course of nerve bundles,which was approximately halfway between the lingula andramus anterior border. The IAN was farthest away from thecortical bone at 70e80% of the course of nerve bundles,which was at approximately the second molar position. Forthe left and right sides, 86.5% (83/96) and 91.7% (88/96) ofthe patients, respectively, had their R point located afterthe point with the minimum of the shortest distances of theouter/buccal edge of the mandibular canal to the innersurface of the buccal cortex.

Tsuji et al15 studied 35 patients with skeletal Class IIIprognathism with symmetry based on transaxial computedtomograms acquired with a slice thickness of 2 mm; they

Figure 6 Receiver operating characteristic (ROC) analyses of the shortest distance of the outer/buccal edge of the mandibularcanal to the inner surface of the buccal cortex measured at the lingula (0%) to detect contact or fusion of the mandibular canal. (A)The left side. (B) The right side. The area under the ROC curve (AUC) is 0.953 (p < 0.001) and 0.977 (p < 0.001) for the left side andright side, respectively. The black dots in the figures are the optimized points with maximum sensitivity þ specificity e 1 (i.e., theYouden index). The white circles in the figures are the points with maximum specificity when the sensitivity is 1.0.

Figure 7 Comparison of the shortest distance from the outer/buccal edge of the mandibular canal to the inner surface of thebuccal cortex measured at the lingula (0%) between mandibular canals with and without contact or fusion. (A) The left side. (B) Theright side.

988 C.-Y. Huang, Y.-F. Liao

reported 16 (22.9%) of 70 rami of the individuals withprognathism had contact or fusion of the mandibular canal.In the current study, 17.2% (11/64) of rami in the normalgroup, 23.4% (15/64) of rami in the retrognathism group,and 21.9% (14/64) of rami in the prognathism group hadcontact or fusion with a slightly higher percentage in thelatter two groups than in the normal group. There was nosignificant difference between the three groups in thecourse of mandibular canal, regardless of the side (i.e., leftor right side) or sex, which indicated that the risk of IANinjury after SSRO is similar between the three groups. Whatis important is that the shortest distance from the outer/buccal edge of the mandibular canal to the inner surface ofthe buccal cortex measured at the lingula can predictcontact or fusion of the mandibular canal. When the

shortest distance at the lingula on the left side was<1.4 mm, the possibility of contact or fusion was 90%. Onthe right side, contact or fusion was present in 81% of pa-tients when the shortest distance was <1.25 mm at thelingula.

Cone-beam CT is commonly used in dentistry, end-odontics, and craniofacial surgery, and it has a high degreeof accuracy with respect to spatial measurement.18,19 Angelet al.26 reviewed the CBCT studies of adult dentate patientsand found that the inferior alveolar canal and foraminawere relatively consistent, regardless of age and sex.Studies have also demonstrated good intra- and interexa-miner reliability.17,20 However, scan setting can affect thevisibility of certain structures.27 Ballrick et al.28 found thatspatial resolution was lower with faster scan times and

Mandibular canal in dentofacial relationships 989

larger voxel sizes, and that a 0.2 mm voxel scan had anaverage spatial resolution of 0.4 mm. Chadwick and Lam29

examined the effect of slice thickness and interslice in-terval on CBCT images of dental implant patients and foundstatistically significant differences in the measured boneheight. In the current study, the slice thickness was 2 mm,which may have impacted the results.

This study is limited by its retrospective nature and therelatively small number of patients. In addition, the caseswere consecutive but they were not randomly selected.Missing teeth, tooth position, or the alignment of mandib-ular wisdom teeth may affect the outer/buccal edge of themandibular canal with respect to the inner surface of thebuccal cortex; this factor was not taken into account. Thisstudy included only Chinese individuals, and therefore theresults may not be applicable to other ethnicities.

In conclusion, the results of this study showed no dif-ference in the IAN canal direction in patients with normaldentition, retrognathism, and prognathism, which suggeststhat no group is more vulnerable to IAN injury during SSRO.The shortest distance from the outer/buccal edge of themandibular canal to the inner surface of the buccal cortex,measured at the lingula, can predict contact or fusion be-tween the mandibular canal and the external cortical bone.During SSRO, surgeons should be very careful at the pointhalfway between the lingula and the anterior ramus borderwhere the IAN is nearest the cortical bone.

References

1. Wittwer G, Adeyemo WL, Beinemann J, Juergens P. Evaluationof risk of injury to the inferior alveolar nerve with classicalsagittal split osteotomy technique and proposed alternativesurgical techniques using computer-assisted surgery. Int J OralMaxillofac Surg 2012;41:79e86.

2. Frydman WL. Nerve injuries. Oral Maxillofac Surg Clin NorthAm 1997;9:207e18.

3. Al-Bishri A, Barghash Z, Rosenquist J, Sunzel B. Neurosensorydisturbance after sagittal split and intraoral vertical osteot-omy: as reported in questionnaires and patients’ records. Int JOral Maxillofac Surg 2005;34:247e51.

4. Schultze-Mosgau S, Krems H, Ott R, Neukam FW. A prospectiveelectromyographic and computer-aided thermal sensitivityassessment of nerve lesions after sagittal split osteotomy andLe Fort I osteotomy. J Oral Maxillofac Surg 2001;59:128e39.

5. Panula K, Finne K, Oikarinen K. Neurosensory deficits afterbilateral sagittal split ramus osteotomy of the mandi-bledinfluence of soft tissue handling medial to the ascendingramus. Int J Oral Maxillofac Surg 2004;33:543e8.

6. Plooij JM, Naphausen MT, Maal TJ, Xi T, Rangel FA,Swennnen G, et al. 3D evaluation of the lingual fracture lineafter a bilateral sagittal split osteotomy of the mandible. Int JOral Maxillofac Surg 2009;38:1244e9.

7. Muto T, Takahashi M, Akizuki K. Evaluation of the mandibularramus fracture line after sagittal split ramus osteotomy using3-dimensional computed tomography. J Oral Maxillofac Surg2012;70:e648e52.

8. Yamauchi K, Takahashi T, Kaneuji T, Nogami S, Yamamoto N,Miyamoto I, et al. Risk factors for neurosensory disturbanceafter bilateral sagittal split osteotomy based on position ofmandibular canal and morphology of mandibular angle. J OralMaxillofac Surg 2012;70:401e6.

9. Yoshioka I, Tanaka T, Khanal A, Habu M, Kito S, Kodama M,et al. Relationship between inferior alveolar nerve canal

position at mandibular second molar in patients with progna-thism and possible occurrence of neurosensory disturbanceafter sagittal split ramus osteotomy. J Oral Maxillofac Surg2010;68:3022e7.

10. Aizenbud D, Ciceu C, Hazan-Molina H, Abu-El-Naaj I. Rela-tionship between inferior alveolar nerve imaging and neuro-sensory impairment following bilateral sagittal split osteotomyin skeletal Class III cases with mandibular prognathism. Int JOral Maxillofac Surg 2012;41:461e8.

11. Kuroyanagi N, Miyachi H, Ochiai S, Kamiya N, Kanazawa T,Nagao T, et al. Prediction of neurosensory alterations aftersagittal split ramus osteotomy. Int J Oral Maxillofac Surg 2013;42:814e22.

12. Yamamoto R, Nakamura A, Ohno K, Michi KI. Relationship ofthe mandibular canal to the lateral cortex of the mandibularramus as a factor in the development of neurosensory distur-bance after bilateral sagittal split osteotomy. J Oral MaxillofacSurg 2002;60:490e5.

13. Kumar Potu B, Jagadeesan S, Bhat KM, Rao Sirasanagandla S.Retromolar foramen and canal: a comprehensive review on itsanatomy and clinical applications. Morphologie 2013;97:31e7.

14. Gowgiel JM. The position and course of the mandibular canal. JOral Implantol 1992;18:383e5.

15. Tsuji Y, Muto T, Kawakami J, Takeda S. Computed tomographicanalysis of the position and course of the mandibular canal:relevance to the sagittal split ramus osteotomy. Int J OralMaxillofac Surg 2005;34:243e6.

16. Kamburo�glu K, Kilic C, Ozen T, Yuksel SP. Measurements ofmandibular canal region obtained by conebeam computed to-mography: a cadaveric study. Oral Surg Oral Med Oral PatholOral Radiol Endod 2009;107:e34e42.

17. Santos Tde S, Gomes AC, de Melo DG, Melo AR, Cavalcante JR,de Araujo LC, et al. Evaluation of reliability and reproducibilityof linear measurements of cone-beam-computed tomography.Indian J Dent Res 2012;23:473e8.

18. Naitoh M, Nakahara K, Suenaga Y, Gotoh K, Kondo S, Ariji E.Comparison between cone-beam and multislice computed to-mography depicting mandibular neurovascular canal struc-tures. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 2010;109:e25e31.

19. Kim TS, Caruso JM, Christensen H, Torabinejad M. A compari-son of cone-beam computed tomography and direct measure-ment in the examination of the mandibular canal and adjacentstructures. J Endod 2010;36:1191e4.

20. Lagravere MO, Low C, Flores-Mir C, Chung R, Carey JP, Heo G,et al. Intraexaminer and interexaminer reliabilities of land-mark identification on digitized lateral cephalograms andformatted 3-dimensional cone-beam computerized tomogra-phy images. Am J Orthod Dentofacial Orthop 2010;137:598e604.

21. Tom WK, Martone CH, Mintz SM. A study of mandibular ramusanatomy and its significance to sagittal split osteotomy. Int JOral Maxillofac Surg 1997;26:176e8.

22. Muto T, Shigeo K, Yamamoto K, Kawakami J. Computedtomography morphology of the mandibular ramus in prog-nathism: effect on the medial osteotomy of the sagittalsplit ramus osteotomy. J Oral Maxillofac Surg 2003;61:89e93.

23. Nagadia R, Tay AB, Chan LL, Chan ES. The spatial location ofthe mandibular canal in Chinese: a CT study. Int J Oral Max-illofac Surg 2011;40:1401e5.

24. Yoshida T, Nagamine T, Kobayashi T, Michimi N, Nakajima T,Sasakura H, et al. Impairment of the inferior alveolar nerveafter sagittal split osteotomy. J Craniomaxillofac Surg 1989;17:271e7.

26. Angel JS, Mincer HH, Chaudhry J, Scarbecz M. Cone-beamcomputed tomography for analyzing variations in inferior

990 C.-Y. Huang, Y.-F. Liao

alveolar canal location in adults in relation to age and sex. JForensic Sci 2011;56:216e9.

27. Hassan BA, Payam J, Juyanda B, van der Stelt P, Wesselink PR.Influence of scan setting selections on root canal visibility withcone beam CT. Dentomaxillofac Radiol 2012;41:645e8.

28. Ballrick JW, Palomo JM, Ruch E, Amberman BD, Hans MG.Image distortion and spatial resolution of a commercially

available cone-beam computed tomography machine. Am JOrthod Dentofacial Orthop 2008;134:573e82.

29. Chadwick JW, Lam EW. The effects of slice thickness andinterslice interval on reconstructed cone beam computedtomographic images. Oral Surg Oral Med Oral Pathol OralRadiol Endod 2010;110:e37e42.