Morphologic Relationship Between the Cranial Base and the Mandible in Patients with Facial Asymmetry and Mandibular Prognathism

ORIGINAL ARTICLE

Morphologic relationship between the cranial base and the mandible in patients with facial asymmetry and mandibular prognathism

Sung-Jin Kim,a Kee-Joon Lee,b Sang-Hwy Lee,c and Hyoung-Seon Baikd Seoul, South Korea

Introduction: This study was conducted to measure the dimensional changes in the cranial base and the mandible in patients with facial asymmetry and mandibular prognathism, and to examine the morphologic relationship between asymmetries of the cranial base and the mandible. Methods: The patients were 60 adults with mandibular prognathism, divided into a symmetry group (menton deviation, \2 mm; n 5 30) and an asymmetry group (menton deviation, .4 mm; n 5 30) according to the degree of menton deviation. Three-dimensional computed tomography scans were obtained with a spiral scanner. Landmarks were designated on the reconstructed 3-dimensional surface models. Linear, angular, and volumetric measurements of the cranial base and mandibular variables were made. Results: In the asymmetry group, the hemi-base, anterior cranial base, and middle cranial base volumes were significantly larger (P \0.01), and crista galli to sphenoid, sphenoid to petrous ridge, anterior clinoid process to petrous ridge, and vomer to petrous ridge lengths were significantly longer (P \0.05) on the nondeviated side than on the deviated side. Menton deviation was significantly correlated with the difference in hemi-base volume, and ramal volume was significantly correlated with the difference in hemi-base and middle cranial base volumes between the nondeviated and deviated sides (P \0.05). Conclusions: In patients with facial asymmetry and mandibular prognathism, cranial base volume increased on the nondeviated side and was also correlated with mandibular asymmetry. (Am J Orthod Dentofacial Orthop 2013;144:330-40)

acial asymmetry is a complex problem that mani- Severt and Profﬁt8 reported that in a group with facial Ffests in inconsistent size, form, and arrangement asymmetry, asymmetry of the upper face was seen in of the facial features on either side of the median only 5% of the patients, whereas 36% showed middle sagittal plane.1 Hereditary and environmental factors third asymmetry and 75% had deviation of the chin. during the fetal, infant, or adolescent periods are known Cheon and Suhr9 and Haraguchi et al10 also reported to contribute to the expression of craniofacial asymme- that asymmetry was found most obviously in the lower try.2 Various etiologic factors have been proposed con- part of the face in the posteroanterior cephalometric ra- cerning facial asymmetry, including unilateral condylar diographs. For these reasons, most studies regarding hyperactivity,3 functional disharmony of the masticatory facial asymmetry have analyzed mandibular asymmetry muscles,4 hemisphere dominance of the brain,4 plagio- using 3-dimensional computed tomography, but there cephaly caused by head posture during early infancy,5,6 are few studies regarding cranial base asymmetry.11-15 and unilateral craniosynostosis.7 Facial asymmetry is often accompanied by varying de- grees of cranial base asymmetry as well as mandibular 16 From the School of Dentistry, Yonsei University, Seoul, South Korea. asymmetry, but cranial base asymmetry is also observed 17,18 aGraduate student, Department of Orthodontics. in patients with no notable facial asymmetry. bAssociate professor, Department of Orthodontics. Additionally, Hayashi19 reported that the morphology cProfessor, Department of Oral and Maxillofacial Surgery. dProfessor, Department of Orthodontics, Institute of Craniofacial Deformity. of the cranial base has an effect on the positions of the All authors have completed and submitted the ICMJE Form for Disclosure of maxilla and the mandible, and that temporomandibular Potential Conﬂicts of Interest, and none were reported. joint position and mandibular shape can also be altered Reprint requests to: Hyoung-Seon Baik, Department of Orthodontics, School of Dentistry, Yonsei University, 134 Shinchon-dong, Seodaemun-gu, Seoul 120- in those with plagiocephaly and an asymmetric cranial 6,7,20 752, South Korea; e-mail, [email protected]. vault and cranial base. However, most previous Submitted, January 2013; revised and accepted, March 2013. studies regarding cranial base asymmetry analyzed 0889-5406/$36.00 Copyright Ó 2013 by the American Association of Orthodontists. plagiocephaly in subjects at early ages, and the effect of http://dx.doi.org/10.1016/j.ajodo.2013.03.024 cranial base asymmetry on mandibular asymmetry is

330 Kim et al 331

still uncertain in adults with facial asymmetries.11,12 The Table I. Patient characteristics in the symmetry and purposes of this study were to measure the dimensions of asymmetry groups the cranial base and the mandible using 3-dimensional computed tomography images of skeletal Class III adults Variable Mean SD Minimum Maximum with mandibular prognathism and to examine the Symmetry group morphologic relationship between asymmetries of the Age (y) 20.8 3.4 18.0 29.0 cranial base and the mandible. ANB ( ) 3.6 2.5 9.9 0.5 Pog to nasion perpendicular 11.7 6.7 5.1 31.4 (mm) MD (mm) 1.0 0.7 0.2 2.0 MATERIAL AND METHODS Asymmetry group Clinical and 2-dimensional radiographic examina- Age (y) 20.9 2.6 18.0 29.0 tions were carried out on patients who visited Yonsei Uni- ANB ( ) 2.6 2.4 8.4 0.1 Pog to nasion perpendicular 9.7 4.5 5.1 21.7 versity Dental Hospital, Seoul, Korea, between 2005 and (mm) 2012. The patients included 60 adults with skeletal Class MD (mm) 8.5 3.4 4.1 16.8 III malocclusion with mandibular prognathism (ANB, \0 Pog to nasion perpendicular, .5.0 mm), no systemic disease, and no degenerative disease of the bone, inferior nasal concha, and zygomatic bone) and temporomandibular joint. Since the chin can greatly cervical bones. The nasal bone was separated from the influence the perception of facial asymmetry, facial frontal bone at the frontonasal suture, and the zygo- asymmetry was defined as menton deviation (MD) from matic bone was separated at the frontozygomatic the midsagittal reference line.10,21 The midsagittal suture. The maxilla was separated from the sphenoid fi reference line was defined as the line passing through bone at the pterygomaxillary ssure, and the zygomatic the crista galli and anterior nasal spine, as bone was separated at the sphenozygomatic suture. In recommended by Grummons and Kappeyne van de the case of the ethmoid bone, only the cribriform plate Coppello.22 The symmetry group included 30 adults (15 and the perpendicular plate were included while the men, 15 women) with an MD from the midsagittal line ethmoid labyrinth was removed. After removal of the less than 2 mm; the asymmetry group included 30 adults facial and cervical bones, the calvaria and the cranial (15 men, 15 women) with an MD from the midsagittal base were separated into a plane connecting the superior line of more than 4 mm (Haraguchi et al10). The patients' point os supraorbitale and both petrous ridge points characteristics in the 2 groups are shown in Table I. (Fig 1, Table II). The 3-dimensional computed tomography scans were Landmarks were designated on the reconstructed 3- obtained before orthognathic surgery using a spiral scan- dimensional surface model, and their positions were fi ner (SOMATOM Sensation 64; Siemens, Erlangen, veri ed on the axial, coronal, and sagittal slices. Land- marks of the cranial base were selected by referring to Germany) with the following settings: gantry angle of 24 11 3 the studies of Captier et al, Kwon et al, Mulliken 0 , 512 512 matrix, 120 kV, 150 mA, 1.0 mm thickness, 25 26 27 and rotation time of 0.5 second. The patients were posi- et al, Moss, and Lee et al, and mandibular landmarks were selected according to the study of You tioned with the Frankfort horizontal plane perpendicular 13 to the floor and the facial midline coinciding with the et al. Landmarks and measurements are given in – long axis of the computed tomography machine. The dig- Tables II and III,andFigs 1 3. The data were measured in increments of 0.01 mm for linear, 0.01 ital imaging and communication in medicine (DICOM) 3 images were created in 1.0-mm slice thicknesses after for angular, and 1 mm for volumetric measurements. scanning. The DICOM data were reconstructed into 3- dimensional images (176 HU threshold value) using the Statistical analysis InVivoDental software program (version 5.1; Anatomage, To account for intraobserver and interobserver errors, San Jose, Calif). the procedure—removing the facial bones and cervical According to the method of You et al,13 the mandible bones and taking measurements—was repeated twice. was separated from the reconstructed image, and the The initial assessment and the reassessment were done teeth above the alveolar bone were removed. Then, as 2 weeks apart by 3 observers (graduate students). The outlined by Enlow,23 the frontal, ethmoid, sphenoid, 2 assessments by each observer were analyzed with the temporal, and occipital bones were all included in the intraclass correlation for intraobserver reliability, and cranial base, which was separated from the recon- the first and second assessments of the 3 observers structed image by removing the facial bones (maxilla, were compared for interobserver reliability. The paired palatine bone, mandible, nasal bone, vomer, lacrimal t test was conducted on measurements performed twice

American Journal of Orthodontics and Dentofacial Orthopedics September 2013 Vol 144 Issue 3 332 Kim et al

Fig 1. Landmarks for segmentation of the cranial base.

Table II. Description of landmarks

Landmark Description Cranial base Cr (crista galli) The most superior edge of the crista galli Cl (anterior clinoid process) Midpoint between the anterior clinoid processes P (petrous ridge) Junction of the superior ridge of the petrous pyramid of the temporal bone and the inner surface of the parietal bone S (sphenoid) The most anterior point of the posterior edge of the lesser wing of the sphenoid Op (opisthion) Midpoint of the posterior arch of foramen magnum V (vomer) The most posterior point of the junction of the sphenoid bone and vomer SOr (supraorbitale) The most superior point of the supraorbital foramen SS (superior point of supraorbitale) The point 10 mm superior to the left supraorbitale Fz (frontozygomatic suture) The most anterior point of the frontozygomatic suture Po (porion) The most superior point of the external auditory meatus Eu (euryon) The most lateral point on either side of the greatest transverse diameter of the head CSOr (contralateral point of SOr) The point where the extension line of SOr-Cl meets the contralateral occipital area Mandible

Consup (condylion superius) The most superior point of the condylar head F (fossa of mandibular foramen) The most inferior point of the fossa of the mandibular foramen

Jlat The most lateral and deepest point of the curvature formed at the junction of the mandibular ramus and body

Jmed The most medial and deepest point of the curvature formed at the junction of the mandibular ramus and body

Gopost (gonion posterius) The most posterior point of the mandibular angle Gomid (gonion midpoint) The midpoint between Gopost and Goinf on the mandibular angle Goinf (gonion inferius) The most inferior point of the mandibular angle MF (mental foramen) The entrance of the mental foramen Me (menton) The most inferior midpoint of the symphysis Pog (pogonion) The most anterior midpoint of the symphysis B (supramentale) The midpoint of the greatest concavity of the anterior border of the symphysis G (genial tubercle) The midpoint of the genial tubercle with a 2-week interval to examine their reproducibility. were not signiﬁcantly different, no differentiation was The method errors were calculated using Dahlberg's for- made for sex. The 2-sample t test was used to compare mula28 (Se 5 O(d2/2n) (d, difference between measure- the measurements between the symmetry and asymme- ments; n, number of measurement pairs). Because the try groups, and the paired t test was also used to 2-sample t test showed that the MD and differences in compare the measurements between the nondeviated the measurements between the male and female subjects and the deviated sides. The Pearson correlation analysis

September 2013 Vol 144 Issue 3 American Journal of Orthodontics and Dentofacial Orthopedics Kim et al 333

Table III. Description of measurements

Measurement Description Cranial base Volume Hemi-base volume The cranial base volume was divided into 2 hemi-base volumes by the plane connecting Cr-Cl-Op Anterior cranial, middle cranial, and posterior cranial volumes The hemi-base volume was divided into anterior, middle, and posterior cranial by the planes connecting Cl-S-V and Cl-P-V Angle Anterior cranial angle The angle formed by the intersection of Cr-Cl and Cl-S (:Cr-Cl-S) Middle cranial angle The angle formed by the intersection of S-Cl and Cl-P (:S-Cl-P) Postcranial angle The angle formed by the intersection of P-Cl and Cl-Op (:P-Cl-Op) Petrous ridge angle The angle formed by the intersection of Cr-Cl and Cl-P (:Cr-Cl-P) Length Cr-S length Distance between Cr and S S-P length Distance between S and P P-Op length Distance between P and Op Cl-S length Distance between Cl and S Cl-P length Distance between Cl and P V-S length Distance between V and S V-P length Distance between V and P Transverse cranial asymmetry SOr (non)-CSOr (dev) length – SOr (dev)-CSOr (non) length CVA (cranial vault asymmetry) Fz (non)-Eu( dev) length – Fz (dev)-Eu (non) length FOA (fronto-orbital asymmetry) Fz (non)-Po (dev) length – Fz (dev)-Po (non) length Mandible Volume Hemi-mandibular volume The mandibular volume was divided into 2 hemi-mandibular volumes by the plane connecting Me-B-G Ramal volume and body volume The hemi-mandibular volume was divided into ramal and body volumes

by the plane connecting Gomid-Jlat-Jmed Length

Condylar unit length Distance between Consup and F Body unit length Distance between F and MF

Non, Nondeviated side; dev, deviated side.

Fig 2. Landmarks and measurements of the mandible. was used to determine correlations between MD and the RESULTS measurements. Statistical evaluations were performed Intraclass correlation coefﬁcients were .908 to .964 ﬁ at the 5% level of signi cance with SPSS for Windows for intraobserver reliability and 0.902 to 0.919 for inter- software (version 18.0; SPSS, Chicago, Ill). observer reliability, indicating high reliability of these

American Journal of Orthodontics and Dentofacial Orthopedics September 2013 Vol 144 Issue 3 334 Kim et al

Fig 3. Landmarks and measurements of the cranial base. measurements. The paired t test showed no statistically was no statistically significant difference in other significant difference between the measurements made measurements of the cranial base (Table IV). 2 weeks later. Among the measurements of the cranial In the asymmetry group, the hemi-base, anterior cra- base, the method errors were 0.56 to 1.68 mm for linear nial base, and middle cranial base volume measurements measurements, 0.85 to 1.95 for angular measure- were significantly greater on the nondeviated side than ments, and 123.57 and 385.25 mm3 for volumetric on the deviated side (P \0.01). The Cr-S, S-P, Cl-P, measurements. Among the mandibular measurements, and V-P lengths were significantly longer on the nonde- the method errors were 0.45 to 1.03 mm for linear mea- viated side than on the deviated side (P \0.05). There surements and 56.70 to 130.58 mm3 for volumetric was no statistically significant difference in angular measurements. measurements of the cranial base (Table V). The asymmetry group showed significantly greater The asymmetry group showed significantly greater differences in hemi-base volume (P\0.01), anterior cra- differences in condylar unit length (P \0.01), body nial base volume (P \0.05), and middle cranial base unit length (P \0.01), hemi-mandibular volume volume (P \0.05) between the nondeviated and devi- (P \0.05), and ramal volume (P \0.01), but there was ated sides compared with the symmetry group. There no statistically significant difference in body volume

September 2013 Vol 144 Issue 3 American Journal of Orthodontics and Dentofacial Orthopedics Kim et al 335

DISCUSSION Table IV. Comparison of differences in measurements of the cranium between the symmetry and asymmetry Traditionally, the cranial base was a 2-dimensional groups (2-sample t test) concept, extending from anterior nasion to posterior basion in lateral cephalograms.29 Furthermore, it is Symmetry Asymmetry difficult to define the border of the cranial base because group group it is actually a 3-dimensional structure composed of the Measurement Mean SD Mean SD P value frontal, ethmoid, sphenoid, temporal, and occipital Volume difference (3103 mm3) 23 y bones. Hence, in this study, these bones were all Hemi-base volume 0.22 2.66 2.50 2.50 0.001 included for volumetric measurements of the cranial Anterior cranial base 0.23 1.30 0.98 1.41 0.036* volume base. To separate the cranial base from the calvaria, a Middle cranial base 0.08 2.93 1.55 2.48 0.024* reproducible plane was needed as close to the floor of volume the cranial vault as possible. A plane connecting both Posterior cranial base 0.07 2.81 0.03 2.44 0.884 P points (prominent points on the border of the middle volume cranial base and posterior cranial base) and supraorbi- Angle difference () Anterior cranial angle 0.13 3.01 1.00 3.30 0.170 tale (located closest to the orbital plate of the frontal Middle cranial angle 0.18 3.22 0.37 3.58 0.534 bone) resulted in the problem of a portion of the orbital Posterior cranial angle 0.01 2.17 0.32 2.23 0.596 plate of the frontal bone being cut off, so SS point was Petrous ridge angle 0.02 2.74 0.62 3.08 0.400 designated 10 mm above the left supraorbitale to Length difference (mm) construct the SSL-PL-PR plane. After designating V Cr-S length 0.17 2.08 0.92 2.24 0.055 fi S-P length 0.57 3.24 1.17 2.70 0.441 point, which is easily identi able and located on the P-Op length 0.29 2.44 0.31 2.89 0.983 lower surface of the body of sphenoid bone, the cranial Cl-S length 0.12 1.94 0.65 3.19 0.265 base was further divided into anterior, middle, and pos- Cl-P length 0.53 3.61 1.56 3.19 0.248 terior cranial bases using the Cl-S-V and Cl-P-V planes, V-S length 0.79 1.22 0.09 2.16 0.130 as reported by Captier et al24 and Kwon et al.11 Finally, V-P length 0.73 3.85 1.61 3.19 0.343 Transverse cranial 0.90 3.66 0.31 3.98 0.552 the cranial base was divided into nondeviated and devi- asymmetry ated sides according to the direction of MD using the CVA 1.33 5.47 2.33 5.98 0.500 Cr-Cl-Op plane.11 FOA 0.80 2.59 0.68 3.70 0.880 The asymmetry group showed significantly greater Difference, Nondeviated side minus deviated side. differences in hemi-base, anterior cranial base, and mid- *P \0.05; yP \0.01. dle cranial base volumes between the nondeviated and deviated sides compared with the symmetry group. (Table VI). In the asymmetry group, hemi-mandibular Also, when comparing the nondeviated and deviated volume and ramal volume were greater on the nondevi- sides in the asymmetry group, the hemi-base, anterior ated side (P \0.01), and condylar unit length and body cranial base, and middle cranial base volumes were all unit length were longer on the nondeviated side significantly greater on the nondeviated side. Moreover, (P \0.01). MD was significantly correlated with the dif- MD was significantly correlated with the difference in ferences in condylar unit length (P \0.01), body unit hemi-base volume in the asymmetry group. These results length (P \0.01), hemi-mandibular volume (P \0.05), suggest that the morphology of the cranial base is altered and ramal volume (P \0.01; Table VII). in patients with facial asymmetry; this contrasts with the In the asymmetry group, MD was significantly corre- findings of Kwon et al11 and Baek et al,12 who reported lated with the difference in hemi-base volume (P\0.05; that the morphology of the cranial base was not related Table V). The difference in ramal volume was signifi- to MD. They examined angular measurements such as cantly correlated with the differences in hemi-base the anterior, middle, and posterior cranial base angles volume and middle cranial base volume (P \0.05; and found no significant difference between the asym- Table VIII). The difference in hemi-base volume was metry and symmetry groups, or between the nondeviated significantly correlated with the differences in Cl-P and deviated sides. In this study, there was also no signif- (P \0.01), V-P (P \0.05), and S-P length (P \0.05), icant difference in the angular and linear measurements whereas the difference in anterior cranial base volume between the asymmetry and symmetry groups, but sig- was significantly correlated with the difference in Cr-S nificant differences were found in hemi-base, anterior length (P \0.05). Furthermore, the difference in middle cranial base, and middle cranial base volumes, indicating cranial base volume was significantly correlated with the the limitations of 2-dimensional measurements in difference in S-P length (P \0.01; Table IX). the evaluation of morphologic changes. Volumetric

American Journal of Orthodontics and Dentofacial Orthopedics September 2013 Vol 144 Issue 3 336 Kim et al

Table V. Comparison of measurements of the cranium between the nondeviated and deviated sides (paired t test), and correlation with MD (Pearson correlation analysis) in the asymmetry group

Nondeviated side Deviated side Difference

Measurement Mean SD Mean SD Mean SD P value Correlation with MD (P value) Volume (3103 mm3) Hemi-base volume 79.64 12.76 77.14 12.78 2.50 2.50 \0.001y 0.393* (0.031) Anterior cranial base volume 16.96 3.44 15.98 3.34 0.98 1.41 0.001y 0.246 (0.190) Middle cranial base volume 33.05 6.52 31.50 6.20 1.55 2.48 0.002y 0.069 (0.718) Posterior cranial base volume 29.63 5.30 29.66 5.45 0.03 2.44 0.945 0.331 (0.074) Angle () Anterior cranial angle 61.58 4.38 60.58 3.55 1.00 3.30 0.108 0.042 (0.827) Middle cranial angle 69.94 4.31 70.31 4.01 0.37 3.58 0.573 0.293 (0.116) Posterior cranial angle 52.23 3.49 52.55 2.89 0.32 2.23 0.442 0.234 (0.213) Petrous ridge angle 131.03 3.80 130.41 2.93 0.62 3.08 0.280 0.283 (0.130) Length (mm) Cr-S length 45.60 2.63 44.67 2.15 0.92 2.24 0.032* 0.175 (0.356) S-P length 74.92 4.73 73.75 5.37 1.17 2.70 0.024* 0.107 (0.573) P-Op length 67.39 3.61 67.08 3.83 0.31 2.89 0.563 0.254 (0.175) Cl-S length 43.75 4.05 43.10 2.90 0.65 3.19 0.271 0.206 (0.275) Cl-P length 77.53 3.83 75.96 4.56 1.56 3.19 0.012* 0.108 (0.570) V-S length 57.62 3.92 56.92 3.64 0.09 2.16 0.071 0.139 (0.463) V-P length 83.29 3.53 81.69 4.79 1.61 3.19 0.010* 0.023 (0.903) Transverse cranial asymmetry 161.07 9.69 161.38 8.90 0.31 3.98 0.671 0.171 (0.365) CVA 147.64 7.68 149.97 6.49 2.33 5.98 0.041* 0.048 (0.802) FOA 128.69 6.86 129.37 7.16 0.68 3.70 0.324 0.171 (0.048)

Difference, Nondeviated side minus deviated side. *P \0.05; yP \0.01.

was no significant difference in body volume, confirming Table VI. Comparison of differences in measurements that both condylar and body units appeared to contribute of the mandible between the symmetry and asymmetry to mandibular asymmetry, with a more crucial role of the groups (2-sample t test) condylar unit, as proposed by You et al.13 Symmetry Asymmetry In the asymmetry group, the differences in hemi-base group group and middle cranial base volumes were correlated with Measurement Mean SD Mean SD P value difference in ramal volume, and the difference in Hemi-mandibular volume 0.22 1.14 1.13 1.93 0.032* hemi-base volume was correlated with MD, indicating difference (3103 mm3) that growth of the cranial base and growth of the Ramal volume difference 0.06 0.67 1.01 1.23 0.001y mandible are related. Craniofacial development was 3 3 (310 mm ) traditionally explained by the functional matrix the- Body volume difference 0.16 1.03 0.11 1.03 0.846 ory30,31 and the counterpart analysis.32 Since the cranial (3103 mm3) Condylar unit length 0.3 1.35 4.78 3.69 \0.001y base is located between the brain and the face, its devel- difference (mm) opment is influenced not only by the neurocranial Body unit length 0.2 1.58 3.33 2.37 \0.001y capsular matrix, but also by the orofacial capsular ma- difference (mm) trix, which can also influence growth of the mandible. Difference, Nondeviated side minus deviated side. Moreover, the mandibular condyle is structurally related *P \0.05; yP \0.01. to the condylar fossa of the temporal bone, thus indicating that the dimensions and placement of 1 part measurements can be useful in determining whether the are closely related to its counterpart. Kwon et al11 re- overgrowth is accompanied by an increase in osteogene- ported that the position of the mandibular condyle is sis, which implies regional hypertrophy. correlated with the position of the mastoid process and The asymmetry group showed significantly greater the petrous ridge angle and might reflect the positional differences in hemi-mandibular volume, ramal volume, change of the petrous ridge because the petrous ridge, condylar unit length, and body unit length between the condylar fossa, and mastoid process are all components 2 sides compared with the symmetry group, but there of the same osseous unit (temporal bone). In addition,

September 2013 Vol 144 Issue 3 American Journal of Orthodontics and Dentofacial Orthopedics Kim et al 337

Table VII. Comparison of measurements of the mandible between the nondeviated and deviated sides (paired t test), and correlation with MD (Pearson correlation analysis) in the asymmetry group

Nondeviated side Deviated side Difference

Measurement Mean SD Mean SD Mean SD P value Correlation with MD (P value) Hemi-mandibular volume (3103 mm3) 33.68 6.29 32.55 6.00 1.13 1.93 0.03y 0.409* (0.025) Ramal volume (3103 mm3) 11.28 2.51 10.26 2.34 1.01 1.23 \0.001y 0.505y (0.004) Body volume (3103 mm3) 22.40 3.99 22.29 3.95 0.11 1.03 0.555 0.161 (0.397) Condylar unit length (mm) 49.79 4.38 45.01 4.96 4.78 3.69 \0.001y 0.514y (0.004) Body unit length (mm) 62.95 4.35 59.62 4.13 3.33 2.37 \0.001y 0.550y (0.002)

Difference, Nondeviated side minus deviated side. *P \0.05; yP \0.01.

Table VIII. Correlation between differences of measurements of the cranium and the mandible in the asymmetry group (Pearson correlation analysis)

Hemi-mandibular Ramal volume Body volume Condylar unit Body unit length Correlation coefﬁcient (P value) volume difference difference difference length difference difference Hemi-base volume difference 0.306 (0.101) 0.388* (0.034) 0.161 (0.397) 0.322 (0.082) 0.299 (0.109) Anterior cranial base volume difference 0.190 (0.316) 0.135 (0.478) 0.193 (0.307) 0.201 (0.286) 0.178 (0.347) Middle cranial base volume difference 0.343 (0.064) 0.422* (0.020) 0.136 (0.474) 0.156 (0.409) 0.125 (0.511) Posterior cranial base volume difference 0.145 (0.443) 0.110 (0.563) 0.140 (0.460) 0.054 (0.775) 0.330 (0.075) Difference, Nondeviated side minus deviated side. *P \0.05.

Table IX. Correlation between differences in volumetric and linear measurements of the cranial base in the asymmetry group (Pearson correlation analysis)

Correlation coefﬁcient Cr-S length S-P length P-Op length Cl-S length Cl-P length V-S length V-P length (P value) difference difference difference difference difference difference difference Hemi-base volume 0.056 (0.768) 0.402* (0.028) 0.465y (0.010) 0.089 (0.639) 0.517y (0.003) 0.249 (0.185) 0.432* (0.017) difference Anterior cranial base volume 0.385* (0.036) 0.142 (0.454) 0.165 (0.384) 0.05 (0.794) 0.114 (0.547) 0.221 (0.240) 0.186 (0.326) difference Middle cranial base volume 0.344 (0.062) 0.581y (0.001) 0099 (0.604) 0.041 (0.828) 0.098 (0.605) 0.062 (0.747) 0.180 (0.342) difference Posterior cranial base volume 0.070 (0.713) 0.098 (0.608) 0.672y \0.001 0.021 (0.914) 0.364* (0.048) 0.189 (0.317) 0.367* (0.046) difference Difference, Nondeviated side minus deviated side. *P \0.05; yP \0.01.

Endo et al33 reported that patients with mandibular growth pattern, which is almost ﬁnished by the age of asymmetry had larger condylar fossae and longer 7 to 8 years, the individual parts of the cranial base condylar processes on the nondeviated side. These re- follow either the neural or the general skeletal growth sults support the ideas that the cranial base and the patterns depending on the location in the base, meaning mandible are structurally related and that their growth that some parts will continue to grow until later stages of increases on the nondeviated side under the inﬂuence adolescence.29,34,35 Our results showed that Cr-S, S-P, of the same orofacial capsular matrix. Cl-P, and V-P lengths on the nondeviated side of the Although cranial base asymmetry appears to be asymmetry group were longer than on the deviated related to mandibular asymmetry, the timing of the side and were correlated with the increases in hemi- development of cranial base asymmetry is uncertain. base, anterior, and middle cranial base volumes, indi- Whereas growth of the calvaria largely depends on cating that the differential transverse and sagittal growth of the brain and thus conforms to the neural growth of both lateral parts of the anterior and middle

American Journal of Orthodontics and Dentofacial Orthopedics September 2013 Vol 144 Issue 3 338 Kim et al

cranial bases contributed to the development of asym- infantile period is a dominant factor in determining the metry. Hoyte36 reported that such transverse and sagittal affected side of deformational plagiocephaly, the pos- growth of the lateral parts of the cranial base continues terior cranial base, which is most frequently influenced until the age of 10 years, and Lang37 reported doubling by deformational plagiocephaly, could be distorted in the length of the petrous bone, from 23 mm in new- regardless of any mandibular asymmetry. This is sup- borns to 37.7 mm at 9 to 11 years to 43.7 mm in adults, ported by our results that show that the average differ- thus contributing to both the transverse and sagittal ence in posterior cranial base volume between the 2 growth of the middle and posterior portions of the sides is close to zero, although with a large standard cranial bases. Therefore, the growth periods of the cra- deviation. nial base and the mandible overlap and might be suscep- Another possible explanation is that since the poste- tible to hereditary and environmental factors, causing rior cranial base has a different embryologic origin, it asymmetric growth at the same time. Moreover, if devel- might have a different genetic growth mechanism opment of the cranial base asymmetry precedes develop- when compared with the anterior or middle cranial ment of the mandibular asymmetry, it could be used as base.43 From an embryologic perspective, the cranial a predictor for mandibular asymmetry in children before base can be divided into parts that undergo endochon- adolescence. Further longitudinal studies using 3- dral ossification and parts that undergo membranous dimensional computed tomography images are needed ossification. The bottom of the cranial base is formed to confirm this hypothesis. by endochondral ossification, during which several carti- In spite of increased hemi-base, anterior, and mid- lage portions are fused to form the chondrocranium. The dle cranial base volumes on the nondeviated sides, prechordal chondrocranium, which lies in front of the there was no significant difference in posterior cranial rostral limit of the notochord, is derived from neural base volumes between the 2 sides. A possible explana- crest cells, and the chordal chondrocranium, which lies tion is that asymmetry caused by deformational plagio- posterior to this limit, arises from the paraxial meso- cephaly during the infantile period conceals the derm.44 The membranous ossification parts, consisting increase in posterior cranial base volume on the nonde- of flat bones adjacent to the calvaria, also develop viated side. Unlike anterior plagiocephaly, which is from neural crest cells, with only the occipital region mainly caused by unilateral craniosynostosis with pre- and posterior parts of the otic capsule arising from the mature fusion of the coronal suture, most occipital pla- paraxial mesoderm.44 Therefore, there is a clear distinc- giocephaly is “positional” or “deformational” and tion between the occipital bone and the posterior parts results from external forces applied to the pliable infant of the temporal bones, which belong to the posterior skull in utero or postnatally.38 The prevalence of defor- cranial base and are derived from the paraxial mesoderm, mational plagiocephaly at 4 months has been recently and the anterior and middle cranial bases, which are reported to be between 19.7% and 48%.39,40 derived from neural crest cells. Although patients with synostotic plagiocephaly were It is difficult to understand the precise pathogenesis excluded in this study, it was not easy to determine of cranial base and mandibular asymmetries based on whether they were affected with deformational the findings in this study because of some limitations. plagiocephaly during the infantile period because of First, the molecular basis of the cranial base and lack of references suggesting criteria for diagnosis of mandibular growth and development were not consid- deformational plagiocephaly in adults. For infants, a ered in this study. Recently, there have been many bilateral difference of over 4 mm in transverse cranial studies regarding the role of specific genes in craniofa- length was defined as “cranial asymmetry,”41 and the cial morphogenesis, such as Bmp,45 Fgf/Fgfr,46 Shh,47 mean difference in this length was 10.8 mm in defor- Sox,9,48 and Dlx,49 which have been reported to play mational plagiocephaly. Mortenson and Steinbok42 important roles in controlling signaling pathways in proposed defining mild to moderate deformity as a cra- craniofacial development. However, little is known nial vault asymmetry (CVA) measurement of 3 mm or about specific molecular biologic mechanisms in the more, and moderate to severe deformity as a measure- development of craniofacial asymmetries. Second, this ment of more than 12 mm. Among the 60 patients of was a cross-sectional study of adult patients. Although this study, 15 had a transverse cranial asymmetry of some correlations were found between cranial base more than 4 mm, 37 had CVA between 3 and 12 and mandibular asymmetries, the cause-and-effect rela- mm, and 3 patients had CVA greater than 12 mm. tionship is not yet clear. A longitudinal study with Although the criteria described above were for infants, 3-dimensional computed tomography imaging will be many patients could be regarded as affected by defor- helpful for detecting and understanding a more precise mational plagiocephaly. Since head posture during the relationship between them.

September 2013 Vol 144 Issue 3 American Journal of Orthodontics and Dentofacial Orthopedics Kim et al 339

CONCLUSIONS 15. Baek C, Paeng JY, Lee JS, Hong J. Morphologic evaluation and classification of facial asymmetry using 3-dimensional computed This study was designed to evaluate the morphologic tomography. J Oral Maxillofac Surg 2012;70:1161-9. relationship between asymmetries of the cranial base and 16. Forsberg CT, Burstone CJ, Hanley KJ. Diagnosis and treatment themandibleinpatientswithfacialasymmetryand planning of skeletal asymmetry with the submental-vertical radio- mandibular prognathism using 3-dimensional computed graph. Am J Orthod 1984;85:224-37. tomography images. The findings showed increased cranial 17. Arnold TG, Anderson GC, Liljemark WF. Cephalometric norms for craniofacial asymmetry using submental-vertical radiographs. base and mandibular volumes on the nondeviated side and Am J Orthod Dentofacial Orthop 1994;106:250-6. showed that these were significantly correlated. Conse- 18. Kim YH, Sato K, Mitani H, Shimizu Y, Kikuchi M. Asymmetry of the quently, when planning a treatment strategy for patients sphenoid bone and its suitability as a reference for analyzing cranio- with facial asymmetry and mandibular prognathism, the facial asymmetry. Am J Orthod Dentofacial Orthop 2003;124:656-62. fact that there could be a cranial base asymmetry affecting 19. Hayashi I. Morphological relationship between the cranial base and dentofacial complex obtained by reconstructive computer mandibular asymmetry should be considered. Therefore, it tomographic images. Eur J Orthod 2003;25:385-91. is necessary to carefully evaluate any asymmetry of the cra- 20. Kane AA, Lo LJ, Vannier MW, Marsh JL. Mandibular dysmorphol- nial base in patients with mandibular asymmetry. ogy in unicoronal synostosis and plagiocephaly without synostosis. Cleft Palate Craniofac J 1996;33:418-23. 21. Ahn JS, Hwang HS. Relationship between perception of facial REFERENCES asymmetry and posteroanterior cephalometric measurements. 1. Peck H, Peck S. A concept of facial esthetics. Angle Orthod 1970; Korean J Orthod 2001;31:489-98. 40:284-318. 22. Grummons DC, Kappeyne van de Coppello MA. A frontal asymme- 2. Pirttiniemi PM. Associations of mandibular and facial asymmetries— try analysis. J Clin Orthod 1987;21:448-65. a review. Am J Orthod Dentofacial Orthop 1994;106:191-200. 23. Enlow DH. The human face: an account of the postnatal growth & 3. Obwegeser HL, Makek MS. Hemimandibular hyperplasia—hemi- development of the craniofacial skeleton. New York: Harper & mandibular elongation. J Maxillofac Surg 1986;14:183-208. Row; 1968. 4. Bjork€ A, Bjork€ L. Artificial deformation and cranio-facial asymmetry 24. Captier G, Leboucq N, Bigorre M, Canovas F, Bonnel F, BonnafeA, in ancient Peruvians. J Dent Res 1964;43:353-62. et al. Plagiocephaly: morphometry of skull base asymmetry. Surg 5. Persing J, James H, Swanson J, Kattwinkel J. Prevention and Radiol Anat 2003;25:226-33. management of positional skull deformities in infants. American 25. Mulliken JB, Vander Woude DL, Hansen M, LaBrie RA, Scott RM. Academy of Pediatrics Committee on Practice and Ambulatory Analysis of posterior plagiocephaly: deformational versus synos- Medicine, Section on Plastic Surgery and Section on Neurological totic. Plast Reconstr Surg 1999;103:371-80. Surgery. Pediatrics 2003;112:199-202. 26. Moss SD. Nonsurgical, nonorthotic treatment of occipital plagio- 6. St John D, Mulliken JB, Kaban LB, Padwa BL. Anthropometric cephaly: what is the natural history of the misshapen neonatal analysis of mandibular asymmetry in infants with deformational head? J Neurosurg 1997;87:667-70. posterior plagiocephaly. J Oral Maxillofac Surg 2002;60:873-7. 27. Lee RP, Teichgraeber JF, Baumgartner JE, Waller AL, English JD, 7. Sakurai A, Hirabayashi S, Sugawara Y, Harii K. Skeletal analysis of Lasky RE, et al. Long-termtreatment effectiveness ofmolding helmet craniofacial asymmetries in plagiocephaly (unilateral coronal therapy in the correction of posterior deformational plagiocephaly: a synostosis). Scand J Plast Reconstr Surg Hand Surg 1998;32:81-9. five-year follow-up. Cleft Palate Craniofac J 2008;45:240-5. 8. Severt TR, Proffit WR. The prevalence of facial asymmetry in the 28. Dahlberg G. Statistical methods for medical and biological dentofacial deformities population at the University of North students. New York: Interscience Publications; 1940. Carolina. Int J Adult Orthod Orthognath Surg 1997;12:171-6. 29. Ford EHR. Growth of the human cranial base. Am J Orthod 1958; 9. Cheon OJ, Suhr CH. A posteroanterior roentgenocephalometric 44:498-506. study of skeletal craniofacial asymmetric patients. Korean J Orthod 30. Moss ML, Salentijn L. The primary role of functional matrices in 1990;20:615-31. facial growth. Am J Orthod 1969;55:566-77. 10. Haraguchi S, Takada K, Yasuda Y. Facial asymmetry in subjects 31. Nepola SR. The intrinsic and extrinsic factors influencing the with skeletal Class III deformity. Angle Orthod 2002;72:28-35. growth and development of the jaws: heredity and functional 11. Kwon TG, Park HS, Ryoo HM, Lee SH. A comparison of craniofacial matrix. Am J Orthod 1969;55:499-505. morphology in patients with and without facial asymmetry—a 32. Enlow DH. Handbook of facial growth. Philadelphia: W. B. three-dimensional analysis with computed tomography. Int Saunders; 1982. J Oral Maxillofac Surg 2006;35:43-8. 33. Endo M, Terajima M, Goto TK, Tokumori K, Takahashi I. Three- 12. Baek SH, Cho IS, Chang YI, Kim MJ. Skeletodental factors affecting dimensional analysis of the temporomandibular joint and fossa- chin point deviation in female patients with Class III malocclusion condyle relationship. Orthodontics (Chic.) 2011;12:210-21. and facial asymmetry: a three-dimensional analysis using 34. Bjork€ A. Cranial base development. Am J Orthod 1955;41: computed tomography. Oral Surg Oral Med Oral Pathol Oral Radiol 198-255. Endod 2007;104:628-39. 35. Arat ZM, Turkkahraman H, English JD, Gallerano RL, Boley JC. 13. You KH, Lee KJ, Lee SH, Baik HS. Three-dimensional computed to- Longitudinal growth changes of the cranial base from puberty to mography analysis of mandibular morphology in patients with adulthood. A comparison of different superimposition methods. facial asymmetry and mandibular prognathism. Am J Orthod Angle Orthod 2010;80:537-44. Dentofacial Orthop 2010;138:540.e1-8; discussion 540-1. 36. Hoyte DA. The cranial base in normal and abnormal skull growth. 14. Kim EJ, Palomo JM, Kim SS, Lim HJ, Lee KM, Hwang HS. Maxillo- Neurosurg Clin N Am 1991;2:515-37. facial characteristics affecting chin deviation between mandibular 37. Lang J. Clinical anatomy of the head. Berlin, Germany: Springer- retrusion and prognathism patients. Angle Orthod 2011;81:988-93. Verlag; 1983.

American Journal of Orthodontics and Dentofacial Orthopedics September 2013 Vol 144 Issue 3 340 Kim et al

38. Robinson S, Proctor M. Diagnosis and management of deforma- 44. Sadler TW. Langman's medical embryology. Philadelphia: Lippin- tional plagiocephaly. J Neurosurg Pediatr 2009;3:284-95. cott Williams & Wilkins; 2010. 39. Hutchison BL, Hutchison LA, Thompson JM, Mitchell EA. Plagioce- 45. Nie X, Luukko K, Kettunen P. BMP signalling in craniofacial devel- phaly and brachycephaly in the first two years of life: a prospective opment. Int J Dev Biol 2006;50:511-21. cohort study. Pediatrics 2004;114:970-80. 46. Marie PJ. Fibroblast growth factor signaling controlling bone 40. Saeed NR, Wall SA, Dhariwal DK. Management of positional formation: an update. Gene 2012;498:1-4. plagiocephaly. Arch Dis Child 2008;93:82-4. 47. Aoto K, Shikata Y, Imai H, Matsumaru D, Tokunaga T, Shioda S, et al. 41. Peitsch WK, Keefer CH, LaBrie RA, Mulliken JB. Incidence of cranial Mouse Shh is required for prechordal plate maintenance during brain asymmetry in healthy newborns. Pediatrics 2002;110:e72. and craniofacial morphogenesis. Dev Biol 2009;327:106-20. 42. Mortenson PA, Steinbok P. Quantifying positional plagiocephaly: 48. Lee YH, Saint-Jeannet JP. Sox9 function in craniofacial develop- reliability and validity of anthropometric measurements. J Cranio- ment and disease. Genesis 2011;49:200-8. fac Surg 2006;17:413-9. 49. Gordon CT, Brinas IM, Rodda FA, Bendall AJ, Farlie PG. Role of Dlx 43. Nie X. Cranial base in craniofacial development: developmental genes in craniofacial morphogenesis: Dlx2 influences skeletal features, influence on facial growth, anomaly, and molecular basis. patterning by inducing ectomesenchymal aggregation in ovo. Acta Odontol Scand 2005;63:127-35. Evol Dev 2010;12:459-73.

September 2013 Vol 144 Issue 3 American Journal of Orthodontics and Dentofacial Orthopedics