Posteroanterior Cephalometry: Craniofacial Frontal Analysis: Joseph G. Ghafari

chapter 23
Posteroanterior
Cephalometry: Craniofacial
Frontal Analysis
Joseph G. Ghafari
Complete Cephalometric poor radiographic technique, concern about additional

4
exposure to radiation, and the relatively low potential of
Analysis PA radiographs, which convey information only on asym-
metry and width of the jaws.
Most facial and radiographic records in orthodontics are Early cephalometric radiologists recognized the
based on the profile. The frontal view of the face, and con- importance of a three-dimensional (3-D) approach to
sequently the posteroanterior (PA) cephalogram, should be study the face and dentition.3 The approach gained
an integral part of facial evaluation, given that we present momentum when nonextraction treatment resurfaced in
1
ourselves to the world face forward. Yet, the number of orthodontics,5 and greater emphasis was placed on the
practicing orthodontists who report routine use of frontal need to include PA cephalograms in the orthodontic
2
cephalometric radiographs has been less than 20%, diagnostic records. The various analyses that
despite the introduction of PA and lateral cephalometry at emerged,6–11 based mostly on linear measurements, are
the same time.3 The low percentage may be a conse- reviewed in this chapter, together with the potential uses
quence of the lack of emphasis given to PA cephalometric and limitations of posteroanterior cephalometry in
evaluation in orthodontic programs, partly because of diffi- orthodontic diagnosis and treatment planning. The latter
culties and limitations encountered in conducting such can be improved only by the development of 3-D repro-
evaluation. These problems include the errors associated duction of the craniofacial anatomy, a process that
with reproducing head posture, identifying landmarks of requires integration of the transverse radiographic
structures that are superimposed or not identifiable with dimension in routine orthodontic evaluation.
267
23 Posteroanterior Cephalometry: Craniofacial Frontal Analysis
Orbital plate
of frontal–
upper limit of 1 1
orbital cavity 2 2
Supraorbital
Planum
2
margin
of sphenoid
Lower limit of A
Nasion pituitary fossa Eu
2
Temporal surface
Arcuate eminence
8 7 a
of greater wing
of sphenoid ACB 7 b
Orbitale
6 5 CO 9 c
Zygion
5 d
int (R)
Registration po 9 8 Frz 10
Zygomatic arch
10 7
Key ridge Infratemporal
6 e 11
surface of Basion
maxilla Zyg 11
Central Max
ANS ray 4 12
10 4 3
Mastoid 3 6
process 13 12
Ma
14
Gonion Gonion
Go
Gauthion
AG
13
Medial sagittal plane Menton
and lateral growth axis
Fig 23-1 Landmarks routinely identified on frontal head radiographs according to Broadbent et al3 (see text for description). Additional common landmarks: (Eu)
eurion; (ACB) anterior cranial base points; (Frz) frontomalare temporale, outer edge of frontozygomatic suture; (Zyg) zygion; (Ma) mastoidale. Other landmarks
shown: (CO) center of orbit; (Max) maxillare; (Go) gonion; (AG) antegonion. Horizontal cranial reference lines include the superior contours of the orbit (a); the
anterior cranial base (b); the center of the orbit (c); the zygomaticofrontal suture (d); the center of the zygomatic arch sectional image (e). The main vertical ref-
erence is the midline plane, defined in many ways (see text), often drawn through crista galli (A).
Frontal Analysis Structures and landmarks

The early evaluation of PA cephalometrics relates to the
The PA cephalogram is taken with the patient's head held Bolton standards, whereby a series of commonly used land-
straight (natural head position1) or slightly down4,6,7 (see marks were identified3 (Fig 23-1):
“Head rotation” below). The plane that intersects the ear
rods, which help stabilize the head, is known as the pori- On the right and left sides of the tracings
onic,12 transporionic,13 or otic14 plane or axis because it 1. External peripheral cranial bone surfaces
presumably intersects the external auditory meati. The 2. Coronal sutures
film-object, film–ear rod, or porion-film3 distance deter- 3. Mastoid processes
mines the amount of magnification of the head structure. 4. Occipital condyles
In early traditional cephalometry, the film holder was 5. Planum sphenoidale and superior surface of the floor
placed to touch the nose, and the percentage of magnifi- of the pituitary fossa
cation was computed and corrected.3 Later, the film–pori- 6. Floor of the nose
onic axis distance could be set at a fixed distance (13 cm12 7. Orbital outline and inferior surface of the orbital plate
or 15 cm15) with corresponding magnification factors (see of the frontal bone
“Errors in frontal cephalometrics”). In digital machines, 8. Oblique line formed by the external surface of the
technological requirements dictate a greater “sensor”- greater wing of the sphenoid in the area of the tempo-
object distance (around 20 cm), leading to enlargement ral fossa
factors of more than 13% that can be corrected in the 9. Arcuate eminence
imaging software. 10. Lateral surface of the frontosphenoidal process of the
zygoma and the zygomatic arch down to and includ-
ing the key ridge
268
Frontal Analysis
Fig 23-2 Posteroanterior cephalometric tracing with landmarks

evaluated in studies by Huertas and Ghafari16 (see chapter 15).
(CO) center of orbit; (J) jugale; (U6C, U6A, L6C, L6A) most lateral
point of crown and most apical point of buccal root of maxillary
and mandibular first molars; (U1A, U1C, L1A, L1C) tip of root
apex and incisal edge of maxillary and mandibular central inci-
sors; (Go) gonion; (AG) antegonion; (Ar) articulare; (Cr) superior
point of crista galli; (ANS) anterior nasal spine; (Me) menton.
Points on the outline of the orbital margin: (S) superior; (L) lat-
eral; (I) inferior; (M) medial.
11. Cross section of the zygomatic arch • Anterior cranial base points (ACB), the margins of the
12. Infratemporal surface of the maxilla in the area of the anterior cranial base determined by the shadow of the
tuberosity, which is seen lateral to the lower outlines of intersection of the frontozygomatic processes with the
the key ridge after the eruption of the permanent first lateral extension of the anterior cranial base floor
molars. • Frontomalare temporale (Frz), the outer edge of the
13. Body of the mandible, the ascending rami, coronoid frontozygomatic suture
processes, and the mandibular condyles (when visible). • Zygion (Zyg), the most laterally situated point on the
Gonion (Go) is the most inferior, posterior, and lateral zygomatic arch
point at the gonial angle of the mandible, and articu- • Mastoidale (Ma), the apex of the mastoid process
lare (Ar) is observed at the intersection of the ramus
and temporal bone. In the midline (see Fig 23-1)
14. Complete dentition or selected dental units 2. Midsagittal suture
4. Basion (when visible)
Given the overlap of teeth and other structures, Huer- 6. Crista galli (most superior point at its intersection with
tas and Ghafari16 limited the definition of dental land- the sphenoid), nasal septum, and the tip of the anteri-
marks to the central incisors and first molars (Fig 23-2). or nasal spine (ANS) (when visible)
They defined the axis of the maxillary and mandibular 13. Menton (Me), the most inferior point on the border of
central incisors as between the tip of the root apex (1A) the mandible, at the symphysis
and the incisal edge (1C), centered mediolaterally. Con- The landmarks used most widely relate to the widths of
sidering the difficulty of tracing the maxillary and the maxilla and mandible, specifically through the distance
mandibular molars, they used buccal landmarks to pro- between these bilateral landmarks :
vide a substitute measure of molar axial inclination,
namely, two points on each of the maxillary and • Jugale (J), at the jugal process, the intersection of the
mandibular first molars: 6C, the most lateral point of the outline of the maxillary tuberosity and the zygomatic
crown convexity, and 6A, the most apical point on the buttress. Some authors use the landmark maxillare
buccal root surface. (Max), the intersection of the lateral contour of the max-
Martin and Saller17 defined several other bilateral land- illary alveolar process and the lower contour of the
marks often used as cranial references (see Fig 23-1): zygomatic process of the maxilla.17 In practical applica-
tion, the landmarks J and Max are nearly coincident.
• Eurion (Eu), the most prominent points on either side of • Antegonion (AG), at the antegonial notch, the lateral
the cranium inferior margin of the antegonial protuberances.
269
A A A
B
A B B
A
D E D
F C
C C G
C c
D
C Nasal
B Ethmoid
Sphenoid
F
a d
Fig 23-3 Correspondence of lateral and frontal structures and landmarks. (a)
A
Sphenoid bone: (A) lesser wings; (B) greater wings; (C) pterygoid processes;
(D) dorsum sella; (E) floor of the hypophyseal fossa; (F) spheno-occipital syn-
B
chondrosis. Greater wing at G is the floor of the middle cranial fossa and coin-
cides with the orbital outline. (b) Zygomatic bones: (A) zygomatic frontal
C
D D
C suture; (B) zygomatic temporal suture; (C) inferior surface of occipital bone;
(D) occipital condyles. (c) Maxillary bones: (A) frontomaxillary sutures; (B)
b pterygomaxillary fissure; (C) alveolar process; (D) palatal surface. (d)
Mandible. (After Moyers.18)
Correspondence of images of basic cranial and facial Evaluation of symmetry

structures between lateral and frontal headfilms is shown
in Fig 23-3.18 Slight asymmetry is normal, but the threshold of clinical
Critical to the evaluation of frontal morphology and significance has not been determined, and its acceptability
symmetry are definitions of reference lines. Horizontal ref- probably depends on the region of asymmetry. A clinically
erence lines connect bilateral landmarks and help deter- unfavorable cant of the occlusal plane may be more signif-
mine vertical asymmetry; vertical (midline) reference lines icant if associated with gingival asymmetry during smile. A
discriminate right and left asymmetries. Horizontal cranial 3-mm deviation of the midline of the chin may be more rel-
reference lines have included the anterior cranial base; the evant in one individual, while a ramus height variance of 3
superior and lateral contours of the orbit or center of the mm may be more significant to the esthetics of another
orbit; the zygomaticofrontal suture or the center of the individual. Studies are not yet available to determine
zygomatic arch sectional image.19 Other horizontal lines means for such variations and their impact on morpholog-
have been used in different regions of the face at the level ic balance.
of the nose, maxilla, and mandible. Asymmetry can be assessed through superimposition of
Vertical references have been described in several ways: right and left sides,18,20 direct horizontal and vertical meas-
via midline anatomic landmarks (crista galli, nasion, ANS, urements, or graphic display in a coordinate system
menton); a perpendicular to a horizontal reference through through the frontal mesh diagram analysis (see chapter 15).
a midpoint of bilateral landmarks; or the best-fit line Moyers18 evaluated mandibular asymmetry by superim-
through a series of midpoints of several pairs of bilateral posing right and left sides after rotating one side over the
landmarks measured directly or derived by the least- other around a vertical axis through crista galli to assess
squares method.19 Reliability of references is addressed in mandibular asymmetry (Fig 23-4). Schmid et al20 conceived
more detail under “Errors in frontal cephalometrics” later in two systems of superimposition of the opposite sides: The
this chapter. first system (Fig 23-5,a) includes mandibular (between
270
Frontal Analysis
Ar Ar
50 53 Go
Go Ar Ar
a Me
45
49 Go
Go
3
Me
a b b
Fig 23-4 Analysis of symmetry. (a) Bilateral superimposition. An arbitrary Fig 23-5 (a) Analysis of symmetry relative to upper cranial structures allows
midsagittal plane is drawn after registration on the shadows of crista galli evaluation of left (blue) structural outlines independent of structural
and its immediate region. After structures on the left side of the head are mandibular asymmetry. (b) Evaluation restricted to the mandibular area
traced, the cephalogram is turned over, and the structures on the right side (between menton [Me], gonion [Go] or antegonion, and articulare [Ar]
are drawn for superimpositional evaluation of asymmetry. (b) Direct meas- point), enables assessment of mandibular structural asymmetry. (After
urements, both horizontal and vertical. (After Moyers.18) Schmid et al.20)
menton, gonion, or antegonion, and articulare point), • J-Cr-AG(L): Same angle on the left side
upper cranial, and craniomandibular areas; the second (Fig • UR6: Angle between the tangent to the buccal surface
23-5,b) is restricted to the mandibular area only. According- of the maxillary right first molar and the line J-J
ly, the degree of symmetry demonstrated with the first • UL6: Corresponding angle for the maxillary left first molar
superimposition could be the result of mandibular dis- • LR6: Angle between the tangent to the buccal surface of
placement with or without structural asymmetry, whereas the mandibular right first molar and the line AG-AG
that observed with the second registration would result • LL6: Corresponding angle for the mandibular left first
from structural asymmetry. molar
• IM(R): Angle between the tangent to the buccal surface
of the maxillary right first molar and the tangent to the
Linear and angular analysis buccal surface of the mandibular right first molar
• IM(L): Corresponding angle on the left side
Angular measurements
In the context of developing a frontal mesh diagram analy- Norms are shown within the context of the second investi-
sis (see chapter 15) and specifically defining a cranial trans- gation (see below, “Transverse Growth and Orthopedic
verse reference, Huertas and Ghafari16 identified bilateral Treatment” and Tables 23-7 to 23-10).
centers of the orbit (CO), representing the geometric cen-
ter of the area defined by tangents to the most superior, lat- Linear measurements
eral, inferior, and medial points on the outline of the orbital Several breadth measurements have been defined17,21:
margin (see Fig 23-2). The authors then used CO and crista
galli (Cr) to develop a series of angular measurements: • Maximal cranial breadth, between right and left eurion
• Anterior cranial base width, between the right and left
• J-CO-AG(R): Angle formed by jugale, geometric center of margins of the anterior cranial base (ACB-ACB)
the orbit, and antegonion on the right side • Upper facial breadth or bifrontozygomatic width, the
• J-CO-AG(L): Same angle on the left side distance between right and left frontomalare temporale
• J-Cr-AG(R): Angle formed by jugale, crista galli, and ante- • Zygomatic breadth or bizygomatic width, between
gonion on the right side right and left zygion
271
Fig 23-6 Evaluation of maxillomandibular relationships according to the Rocky Mountain analysis developed by Ricketts.7 R and L indicate right and left for land-
marks antegonion (AG and GA, respectively), jugale (J), and bilateral points on the medial margins of the zygomaticofrontal suture, at the intersection of the orbit
(Z). The frontolateral facial lines (FL) are two lateral lines constructed from ZR and ZL to points AG and GA. (1) The maxillomandibular width differential is a meas-
urement, along the J-J line, of the distance between right and left jugale and the frontolateral facial line. The average distance is 10 ± 1.5 mm. The summed dif-
ference from each side depicts the total transverse deficiency. This method demonstrates differences in deficiency on one side or the other but not in which arch
the discrepancy is located. (2) The maxillomandibular transverse differential index is the difference between (a) the expected (normal values are for Caucasians)
maxillomandibular difference, which is defined as the age-appropriate expected AG-AG distance minus the expected J-J measurement, and (b) the actual max-
illomandibular difference, ie, the actual AG-AG distance minus the actual J-J measurement. Nearly 5 mm around the difference is allowed (for ages 15 years and
older) above the expected difference to delineate diagnosis of severe discrepancy and need for maxillary orthopedic expansion. (Skull drawing adapted from Fai-
gin.23)
• Mastoid breadth or bimastoid width, between right and • CO-CO; distance between the geometric centers of the
left mastoidale orbits
• Maxillary base breadth, between the bilateral landmarks • L6-6A, distance between the apices of the distobuccal
maxillare, practically similar to the distance J-J roots of the mandibular first molars
• Gonial breadth or bigonial width. Ricketts7 introduced as • L6-6C, distance between the most buccal points of the
a substitute for this measurement the distance AG-AG crowns of the mandibular first molars
• U6-6A, distance between the apices of the distobuccal
In addition to the maxillary (J-J) and mandibular roots of the maxillary first molars
(AG-AG) widths, Huertas and Ghafari16 calculated these dis- • U6-6C, distance between the most buccal points of the
tances: crowns of the maxillary first molars
272
Frontal Analysis
Presumably, the relationship between the widths of the regarding the side deviant from the “norm.” Validity of refer-
maxillary and mandibular skeletal bases is the most critical ence lines must be related not only to (external) errors per-
information sought from the PA record. Among several tinent to the cephalometric method (ie, object-film dis-
4,6–11,22 6,7
analyses, Ricketts’s method (also known as the tance, head rotation) but also to two (internal or inherent)
Rocky Mountain analysis) appears to be used most widely factors of variability: (1) actual variation in landmark location
(Fig 23-6). In this method, which provides normative values and (2) error of identification of landmarks because of the
for different ages, the maxillomandibular width difference clarity (recognition potential) of their corresponding
is computed by determining the normal “expected” differ- anatomic structures in the PA record. Accordingly, certain
ence for the age evaluated, then the “actual” maxillo- references are more reliable than others in the determina-
mandibular differential; the balance between expected tion of asymmetry.
and actual differences reflects the existing discrepancy. The reliability of frontal cephalometry has been investi-
Additional information on the Ricketts analysis is presented gated in many studies,12,14,19,28–37 although not to the extent
in chapter 8. of lateral radiography. The variability of the PA radiograph
In clinical application, the standard deviation (SD) is involves the influence on the anatomic images of film-
taken into consideration for therapeutic decisions. Given a object distance, head angulation, and associated differential
difference of 24 mm instead of 19 mm between the jaws (in errors of magnification at various levels of the headfilm. Con-
an adult), an SD of 5 mm serves as a margin of normalcy sequently, references and measurements may be affected.
that may preclude the widening of a relatively narrow max-
illa in the absence of a posterior crossbite or an increased Film-object distance
buccal corridor (ie, the distance between the buccal teeth Ghafari et al12 investigated the effect of varying film-object
and the lip commissure) that may detract from facial attrac- distance (film–ear rod or film–porionic axis distance [FPD])
tiveness.24 on measurements of distances between bilateral land-
Cortella et al25 provided different data based on the marks. They subjected human skulls to PA radiographic
Bolton longitudinal data (see below, “Transverse Growth exposure at the FPDs of 11, 12, 13, and 14 cm. No clinically
and Orthopedic Treatment” and Tables 23-3 and 23-4). significant (ie, > 1 mm) differences existed between meas-
urements of distances on the skulls and on the headfilms.
Ghafari et al12 also recorded the FPD in 59 human adults
Proportionate analysis (age range, 10 to 45 years; mean, 17 years) when their
heads were positioned in the cephalostat for a PA radi-
In several studies, Ghafari and associates12,24,25 demonstrat- ograph with the FH aligned parallel to the floor and the film
ed that ratios of maxillary width (J-J) to mandibular width cassette lightly touching the nose. The mean FPD was
(AG-AG) may be more suitable than distances for evaluat- 11.53 ± 0.95 cm (minimum, 9.8 cm; maximum, 14.3 cm); the
ing the relation between the jaws. Other authors have also majority (95%) were within a range of 10 to 13 cm, and only
used proportions between frontal structures.11 5% were greater than 13 cm (Fig 23-7).
The power of proportionality is best illustrated in the On the basis of this result and the finding that transverse
mesh diagram analysis (see chapter 15). Originally evaluation should not be affected significantly if the dis-
designed for lateral cephalometry by Moorrees and his tance increased to the maximum observed (14.3 cm), Gha-
associates,1 the transverse mesh diagram followed the fari et al12 suggested that an FPD of 13 cm could be adopt-
same basic principles, namely, the generation of an individ- ed as a practical standard until a universal standard can be
ualized norm on the basis of the patient’s upper facial agreed upon. In European centers,15 an FPD of 15 cm is
height and width, proportionate evaluation of structures, fixed in the cephalostat. This distance should accommo-
simultaneous illustration of the vertical and transverse loca- date all sizes of heads, but its adoption as a universal stan-
tion of landmarks in a coordinate system, and a graphic dis- dard would require a study of large samples of subjects and
play and appreciation of the relationships among essential skulls, remembering that the minimum object-film dis-
facial components needed for orthodontic diagnosis. tance that is practical should be used to minimize enlarge-
ment of the radiographic image.3,38 To this end, in the
Bolton studies Broadbent et al3 positioned the head with
Errors the nose touching the film cassette and subsequently cor-
rected for the magnification at that distance.
Errors in lateral cephalometry are reduced, for better or In digital cephalometry, the “sensor”-object distance is
worse, by averaging asymmetric outlines or location of greater than 15 cm because of the physical setup of the
landmarks.26,27 This strategy may not be optimal for frontal machine and sensors. Instead of the enlargement factor of
cephalometry, at least not without clinical judgment 8.5% with an FPD of 13 cm, or nearly 10% with a distance of
273
Fig 23-7 Porionic axis–film distance. More than 95% of subjects had a dis- Fig 23-8 Geometric principle of similar triangles for correction of magnifica-
tance of 13 cm or less. (From Ghafari et al.12) tion on PA headfilm. (A) Anode; (AB) distance from anode to transporionic
axis; (AC) distance from anode to film; (DE) actual width between facial land-
marks; (FG) measurement of projection of DE on headfilm; (X) corrected dis-
tance between landmark and transporionic axis as measured on lateral
cephalogram of same subject. (From Hsiao et al.13)
15 cm, magnification with digital PA radiographs is about

13.5% to 14%. Since the images are captured directly on the DE = FG ⫻ [(AB ± X)/AC]
computer, imaging software can reduce the enlargement to
a ratio of 1:1 correspondence between object and image. In where DE is the actual width between landmarks on the PA
this instance, available norms should be corrected to factor cephalogram; FG is the linear measurement between land-
out the amount of magnification for more accurate evalua- marks on the PA cephalogram; AB is the anode–transpori-
tion (see below, “Transverse Growth and Orthopedic Treat- onic axis distance; X is the corrected distance from the
ment”). Newer technology, digital or 3-D computerized landmark to the transporionic axis as measured on the lat-
tomography (CT) that eventually will become the standard eral cephalogram of the same subject; and AC is the anode-
imaging method for orthodontics, would eliminate the pres- film distance.
ent attention to magnification errors but also require adjust-
ment of the norms developed over 75 years of traditional Head rotation
cephalometry. Until then, researchers and clinicians should Ghafari et al12 studied the effect of head angulation at +5
maintain a keen interest in error evaluation during a transi- degrees and –5 degrees from FH on measurements of dis-
tional period in which phasing into newer technology shall tances between bilateral landmarks on human skulls and
depend on more refinement of the technologies and associ- on headfilms. Transverse measurements were not statisti-
ated cost issues. cally significant within the investigated 10-degree range.
The advantage of fixing the FPD at the same distance In other studies of geometric variations in head position,
across ages is reflected in longitudinal evaluations of same a change within a 5-degree up-down or left-right rotation
individuals. Yet regional superimpositions are not as com- on width measurements was deemed negligible.28,39 How-
mon or essential as with lateral headfilms. Superimposition ever, Yoon et al32 had a more detailed analysis of left-right
on the anterior cranial base, maxillary base, or mandibular changes. In a study of projection errors of PA cephalograms,
structures would abound with errors on PA cephalograms, they rotated skulls around the vertical z-axis, from 0 degrees
whereas it is more reproducible on lateral records. to ± 10 degrees at 1 degree intervals to the left and right
Hsiao et al13 suggested a method of correcting width sides. The ordinate values were almost unchanged. Most of
measurements from PA cephalograms (Fig 23-8). Including the abscissa values showed statistically significant differ-
a measure from the lateral cephalogram, the correction ences for each rotational angle, effecting an average change
equation is: in the landmarks from –0.19 mm (mastoid process) to 1.69
274
Frontal Analysis
Fig 23-9 Head tilt affects vertical measure-

ments on PA radiograph. All three PA
cephalograms were taken at the same pori-
onic plane–film distance (14 cm). (a) Head
inclined upward by 5 degrees shows shorten-
ing of lower face (ramus, mandibular body,
teeth) and midface (nasal cavity) structures
relative to straight position (b). Less overlap
of structures is seen in the orbital area and
other structures above the porionic plane. (b)
Head held straight (parallel to FH). (c) Head
tilted downward by 5 degrees leads to elon-
gation of lower and midfacial structures. a b c
More overlap of structures is observed above
the porionic axis. (From skulls used in study
by Ghafari et al.12)
Fig 23-10 Cephalic and facial indices. The

breadth-length ratio of the cranium is less
than 76% for dolicocephalic patterns and
Cephalic indices
greater than 81% for brachycephalic pat-
terns. A mesocephalic cranium has a value
in between.40 Prosopy is determined by the
ratio of the distance ophyrion-gnathion
(head height) divided by the bizygomatic Dolichocephalic Mesocephalic Brachycephalic
width ⫻ 100. Leptoprosopic (narrow) faces
have an index greater than 104%; the index
for euryprosopic (broad) faces is less than
97%; mesoprosopic faces are in between.
(After Broadbent et al.3) Facial indices
Leptoprosopic Mesoprosopic Euryprosopic
mm (ANS); the widest changes (≥ 1.5 mm) occurred with since an upward tilt leads to shortening of the lower face
ANS, crista galli (1.61 mm), the nasal cavity (1.59 mm), and image and overlap of its structures with those of the mid-
menton (1.55 mm), all of which are midline-related struc- dle face (see chapter 3). Since lower face height differs with
tures. Landmarks anterior to the rotation axis displaced in facial type (Fig 23-10), the amount of downward tip during
the direction of head rotation, and those posterior to the the taking of the radiograph may depend on the facial
axis displaced in an opposite direction. Such rotations can type. Less inclination is required in a leptoprosopic or
occur if the head is not stable in the cephalostat or if one ear dolichocephalic pattern than in a euryprosopic or brachy-
is forward of the other in the anteroposterior plane. cephalic pattern. Such relationship has not been investigat-
Variation in the tilt of the head relative to the film affects ed. In the classic Bolton studies,4 Broadbent et al3 corrected
the vertical measures more significantly than the transverse for the vertical variance that occurred with the vertical
measures (Fig 23-9). Often, a slight downward tip of the head rotation with the use of an “orientator,” a device pro-
face (about 5 degrees) allows clearer evaluation of the duced for the orientation of the lateral and frontal films to
lower face.4,6 each other. With two x-ray sources used in the Broadbent
Therefore, when positioning the patient in the cephalo- cephalostat,3 one lateral and one posteroanterior, which
stat, the patient’s head must be straight14 to slightly down,4 allowed for both headfilms to be taken for the same head
275
Table 23-1 Comparison of skull and cephalometric linear measurements (n = 13 for each record) (From Chidiac et al.26)
Mean SD P* % difference/
Intraclass r Pearson r difference (mm) difference (mm) difference skull†
Sagittal
Co-Pog 0.73c 0.81b 1.06 4.20 .38 0.94
Go-Gn 0.04n 0.85b -8.32 4.64 < .0001 –10.29
Vertical (lateral view)
N-Me 0.47d 0.99a 9.55 0.90 < .0001 9.02
A-B 0.74b 0.89a 2.74 3.64 .02 7.45
N-UI 0.46d 0.96a 5.91 1.36 < .0001 8.31
Me-LI 0.69c 0.97a 3.48 0.70 < .0001 8.86
Transverse
Or-Ol 0.52d 0.80c –0.57 2.76 .471 –2.12
J-J 0.79b 0.79c 1.14 1.78 .04 1.83
AG-AG 0.46d 0.94a 3.55 2.92 .0009 4.42
(a) P < .0001; (b) P = .0002 to .0008; (c) P = .002 to .006; (d) P = .02 to .05.
(Co) Condylion; (Pog) pogonion; (Go) gonion; (Gn) gnathion; (N) nasion; (Me) menton; (A) point A; (B) point B; (UI) maxillary central incisor; (LI) mandibular central incisor; (O) orbitale right (r); orbitale left (l); (J) jugale; (AG) ante-
gonion.
*Paired t test.
†Percentage of distortion of anatomic measures (skull) on cephalometric view. Percentage distortion computed as mean difference between cephalogram and skull, divided by measurement on skull.
position, their correction was easier than with the common overlooked but must be considered. The clinical implica-
one-source x-ray cephalostat, which requires patient repo- tions of such errors were evaluated in a comparison of PA
sitioning. With new 3-D CT imaging, the problems induced cephalograms of human skulls to the corresponding skull
by head rotation should be significantly reduced if not anatomy.26 The range of error may be deemed slight to
eliminated. negligible for the regular cephalometric assessment and
clinical use. The level of cephalometric distortion gradually
Asymmetric ears and head position decreased for structures closer to the film, which was
Asymmetric ears, either vertically or posteriorly, can affect placed 13 cm anterior to the transporionic plane. The dis-
visualization of the midline; thus, the alignment of a head tortion level varied at different planes; the mandibular
in a cephalostat with two ear rods fixed at the same plane width (AG-AG) was subject to more than twice (4.42%) the
results in head rotation, leading to misinterpretation of distortion of the maxillary width (J-J, 1.83%) (Table 23-1).26
position and symmetry of facial structures that further This finding suggests that a diagnosis of discrepancy
compromises the analysis of symmetry.1 The vertical and PA between maxillary and mandibular widths exaggerates the
locations of the ears should be evaluated clinically, particu- difference by about 2.5% the value of AG-AG, which
larly when a cephalometric asymmetry is evident but a clin- amounts to nearly 2 mm for a mandibular width of 86 mm.
ical asymmetry is not observed. Miyashita41 recommends Grayson et al suggested evaluating craniofacial asym-
identifying the sources of asymmetric images following metry by multiplane analysis.42 They reasoned that features
specific guidelines. of the midline may be described at different depths and
One potential solution is the allowance for one of the ear illustrated their premise through a patient with craniofacial
rods to be removed upon clinical diagnosis of significant microsomia. By integrating observations from various lev-
asymmetry between the external auditory meati, while els of lateral and frontal headfilms, they described progres-
maintaining the midsagittal plane perpendicular to the film sive midline deviation from the anterior to the posterior
holder or sensor (digital machines). Certain machines may aspect of the head (Fig 23-11).
not operate unless the second rod is in a downward position The layered depths lie between the nose and occiput, a
facing the ear. In this instance, the manufacturers should distance estimated at >20 cm ± 6.7 in males and >19 cm ±
allow for the ear rod to be “plugged out” of the ear holder. 6.8 in females (ages 19 to 25 years).43 Comparatively, lateral
cephalograms project images of nearly identical right and
Anatomic level of evaluation left structures between the ears. Each of these areas is
Cephalometric errors of magnification that result from the approximately 50% of the average head width (distance
projection of a 3-D head on a cephalometric film are often euryon-euryon), about 7.5 cm between ages 19 to 25
276
Frontal Analysis
3 2 1
a
3
2 1
Msi Mce
Si
Mz
Mc Mp
Md
Mx Mi
Mm
Mg
Mgo Mf
d c b
Fig 23-11 Tracings on separate acetate sheets are made on the PA headfilm depicting three planes corresponding to structures shown on the lateral cephalogram
(a). (b) Structures corresponding to plane 1: orbital rims (Mce), pyriform aperture (Mp), midpoint between maxillary and mandibular incisors (Mi), inferior border
of the symphysis at gnathion (Mg). (c) Structures at the level of plane 2: intersection of shadows of greater and lesser wings of the sphenoid (Msi), the most later-
al section of the zygomatic arch (Mz), coronoid process (Mc), maxillare (Mx), body of the mandible at the mental foramina (Mf). (d) Structures at the depth of plane
3: heads of mandibular condyles (Md), innermost inferior points on the mastoid processes (Mm), gonions (Mgo). Asymmetry is more severe posteriorly than ante-
riorly. In (b), (c), and (d), M refers to the midpoint between bilateral landmarks. (Adapted from Grayson et al.42)
years,43 an estimate compatible with the Bolton maximum Critical to the PA record are the landmarks that con-
midsagittal line to film distance of 9 cm.3 tribute to the accurate definition of the patient’s facial mid-
line as the origin of analysis. Ideally, if the head is in natural
Reliability and limitations of references head position, the midline would simply be drawn as a per-
Several limitations challenge the development of trans- pendicular to the “true” horizontal (see chapter 15). The
verse analyses from PA headfilms: (1) Asymmetry is a gener- midline is commonly drawn through Cr-ANS; therefore, it is
al characteristic of human faces; (2) the midline, which subject to deviation from the “true” midline by Cr, ANS, or
must be the origin for measurements, is not always easily both. Presumably, nasion may also fall off this determina-
identified; (3) the alignment of a head with asymmetric ears tion. Crista galli on PA headfilms is the closest identifiable
using a cephalostat with two ear rods results in head rota- landmark to nasion on lateral cephalograms. Pending
tion and consequently an artificial distortion of facial char- focused research in this area, crista galli seems less variable
acteristics. As for lateral cephalometric analyses,27 reference than nasion on PA films.
planes and landmarks must be evaluated for appropriate While the contour of crista galli may provide the image
utilization because they are variable and cannot replace of a “channel” to draw the vertical, the landmark Cr is usual-
judgment. ly readily seen at the intersection of the crista galli with the
277
Table 23-2 Errors (mm) of landmarks in horizontal (x) and vertical (y) directions: Summary of published studies
Intra-examiner Inter-examiner (n = 4)*
Landmark x* y* x† y† x y
Cr 0.52 2.07 — — 0.67 2.17
ANS 0.25 0.37 1.12 0.62 0.45 1.20
Me 0.69 0.20 0.42 1.08 0.96 0.72
J† (or M*) 0.77 0.93 0.74 0.54 2.60 3.06
AG 0.42 0.39 0.83 0.64 0.64 0.64
*From Major et al.31
†From El-Mangoury et al.29
(Cr) Crista galli; (ANS) anterior nasal spine; (Me) menton; (J) jugale; (M) deepest point on curvature of maxillary malar process (yields information about maxillary width comparable to that provided by jugale at the jugular process);
(AG) antegonion.
image of the sphenoid bone, unless the nasal bone is ANS position tends to be altered in facial asymmetry that
superimposed on crista galli, an occurrence related to the involves the maxilla. The most valid vertical lines with R2
head being tilted back when the radiograph is taken.41 values greater than or equal to 0.966 did not include mid-
Maintaining the head in natural head position44 or slightly line points. They were perpendicular lines to the horizon-
down4,6 in the cephalostat should minimize this incidence, tal lines that connected the following bilateral points:
as well as the potential distortion and misinterpretation of intersection of zygomaticofrontal suture with lateral
spatial position of structures from significant backward or orbital margin (ZF) and the midpoint of the inferior, lateral,
forward head tilt. and medial orbital margins (OI, OL, and OM, respectively).
ANS is identified at the intersection of the two halves of Such midline perpendicular references are commonly
the maxilla in the midline, below the floor of the nasal cav- used in the study of PA symmetry in patients with cleft
ity. Head positioning influences the identification of ANS. If palate, where maxillary dysmorphology precludes the use
no horizontal or vertical head rotation affected the of midline structures for midline vertical reference
cephalogram, and the image of ANS is not evident, a trans- lines.45–47 Trpkova et al’s findings19 were consistent with
fer of the distance between (the level of ) Cr and ANS from Yoon et al’s findings32 that Cr and ANS changed significant-
the lateral film provides a working compromise. Another ly with right and left rotation of the heads in increments of
accepted compromise to minimize error of identification is 1 degree.
to average the discrepancy between Cr and ANS when no Considering these findings and the original and com-
clear-cut definition can be made on which is closer to the mon use of midline vertical references connecting crista
midline. galli and/or nasion to ANS to draw frontal cephalometric
In a dry skull model subjected to 30 asymmetric posi- norms, and since Cr remains a fairly identifiable and reliable
tions of the maxillomandibular complex, Trpkova et al19 point, Cr may be connected to a midpoint of the more reli-
determined that 10 horizontal lines (nine connecting nine able bilateral distances (ZF or orbital points) to obtain the
bilateral cranial landmarks and one reflecting the best fit of midline vertical reference. In all instances, as for lateral
these landmarks) indicated excellent agreement between cephalometry, reliability of the references used must be
cephalometric and direct measurements, as gauged by the checked in every individual and proper judgment
adjusted R2 values being close to a correspondence of 1:1, employed, specifically when cranial asymmetry exists.
ranging between 0.94 and 0.97. Conversely, of 15 vertical Some authors have investigated the validity of PA
references that included either two anatomic landmarks in measurements by comparing them to corresponding
the midline, or perpendicular lines to the midpoint of bilat- measurements of skulls,12,26,48 and others by cross-referencing
eral analog points, as well as one best-fit line connecting landmarks on lateral cephalograms.49,50 Investigations of the
averages of all cranial bilateral points, only 10 accurately accuracy of landmark identification suggest that landmarks
represented transverse asymmetry. with the least amount of variation (< 1.5 mm) should be
Vertical lines determined between two of four midline considered for cephalometric analysis.27,29,30,51,52 Studies of
landmarks (crista galli, nasion [N], ANS, and menton) had identification errors in PA cephalometry show variable
adjusted R2 values below 0.9 (N-Me: 0.82; Cr-Me: 0.79; Cr-N: ranges of error.29,30 The errors for ANS and the horizontal
0.70; N-ANS: 0.08; Cr-ANS: 0.06). Specifically the lines con- location of Cr were less than 1 mm (Table 23-2), but about 2
necting Cr or N to ANS were not valid, as apparently the mm for the vertical location of Cr. Major et al30 defined the
278
Transverse Growth and Orthopedic Treatment
landmark Cr at the geometric center of the crista galli, dent mandibular asymmetry, leading to a final number of
apparently similar to the definition by Huertas and Ghafari.16 35 subjects. Because the data were not strictly longitudinal
However, the latter authors’ location of Cr seems more spe- between the ages of 5 and 18 years—the interval consid-
cific (thus, possibly subject to less error) because the image ered for this investigation—the total number of subjects
of the sphenoid helped identify the landmark. for any given age ranged between 22 and 34.
Available from studies of identification errors, and list- The distances J-J and AG-AG, measured on tracings of
ed in Table 23-2, are data on landmarks critical for the eval- the cephalograms, were adjusted by subtracting the per-
uation of the lower facial height (menton) and width of the centage of enlargement computed on the basis of the dis-
jaws (jugale, antegonion). In clinical practice, most ortho- tance between the porionic axis and the film. As this dis-
dontists limit the transverse analysis to the difference tance increased, the enlargement was greater. The means
between maxillary and mandibular widths and use this dif- and SDs for the distances J-J, AG-AG, and the correspon-
ference as the basis for treatment planning because of ding differences ([AG-AG]–[J-J]) are presented for the cor-
overarching therapeutic limitations. The errors related to J- rected (Table 23-3) and radiographically enlarged (Table
J and AG-AG are within 1 mm in various studies.12,16,25,29,30 23-4) measurements. The values in Table 23-4 are shown for
Upcoming 3-D CT technology should facilitate recog- comparison with published normative data that are not
4,6–11,53
nition of anatomic structures and thus minimize or even- corrected for enlargement. Since the enlargement
tually eliminate limitations of reference lines, although factors varied at the different age intervals (Table 23-5), the
sound judgment on scientific validity based on variability differences between AG-AG and J-J deviated correspond-
of these references remains important. ingly from the differences calculated for the nonenlarged
measurements.
The mandibular width AG-AG, which is always greater
than J-J, was more affected by the radiographic enlarge-
Transverse Growth and ment, but the ratios of J-J:AG-AG were not different
Orthopedic Treatment between enlarged and corrected measurements. The
development of mandibular width appeared similar in
boys and girls until ages 11 to 12 years (Fig 23-12a). There-
after, the two groups diverged; the difference was statisti-
Transverse development of the jaws cally significant at age 16 years (P < .05). The maxillary width
Differences between the norms given by various studies (Fig 23-12b) followed the same pattern with statistically sig-
warranted the establishment of standards based on longi- nificant differences between boys and girls at ages 17 and
tudinal records collected with methodical rigor. Cortella et 18 years (P < .05).
25 3
al used the Bolton-Brush material as the basis for gener-
ating normative data. Recognized as a controlled longitudi- Inference on growth
nal record of growing children, the Bolton-Brush collection The difference in development of maxillary and particular-
included serial cephalometric headfilms of nearly 5,000 ly mandibular width between girls and boys (see Fig 23-12)
individuals, taken on a 6-month basis in the first 4 years of reflects similar trends for the development of mandibular
54 55
life and thereafter on an annual basis to young adulthood. length and even intermolar distance. Facial growth has
For each participant, the records included lateral and PA been reported to end first in width, then in length, and
56
cephalograms, hand-wrist radiographs, study casts, and finally in height. The present data reveal that increases in
measurements of height and weight. The distance velocity of transverse growth occur at ages that coincide,
between the x-ray tube and the porionic axis was fixed at 5 on average, with the timing of the adolescent growth
feet (1.524 m). The film was placed close to the nose, lead- spurt: around age 11.5 years in girls and 13.5 to 14 years in
57
ing to a different enlargement factor for each radiograph. boys. Differences in body height between boys and girls
58
The film–porionic plane distances were recorded to com- also have been noted after age 13 years.
pute and correct for the enlargement. The increased radiographic J-J distance in males
The sample selected by the Bolton-Brush Growth Center between ages 10 and 18 years (3.3 mm) is similar to Björk’s
consisted of the records of 36 subjects (18 females and 18 findings on average growth in maxillary width (3.0 mm) as
males). The selection criteria for the lateral cephalograms measured on serial radiographs between posterior
59
included the availability of longitudinal records, “excellent implants in boys. Development of facial width, particular-
static occlusion” on study casts, good health, and estheti- ly mandibular width in boys, continues beyond the spurt
cally favorable faces (termed Bolton faces) that conformed periods in a pattern similar to that seen in facial length and
54
to the statistically derived means of craniofacial measure- height. The differential growth between maxilla and
3
ments. One male subject was excluded because of evi- mandible also seems to be similar for all planes, since the
279
Table 23-3 Mean cephalometric values corrected for radiographic enlargement (From Cortella et al.25)
Age AG-AG (mm) J-J (mm) Difference (mm) Ratio (%)
(y) n Mean SD Mean SD Mean SD Mean SD
5 29 65.7 2.7 51.5 2.6 14.2 2.3 78.5 3.1
6 28 67.5 2.8 53.0 2.2 14.5 2.2 78.6 2.8
7 29 68.6 3.1 53.8 2.0 14.8 2.7 78.6 3.2
8 31 70.1 3.1 55.1 2.2 15.0 2.7 78.7 3.2
9 33 71.9 3.2 56.6 2.3 15.4 2.8 78.7 3.2
10 34 73.1 3.1 57.3 2.7 15.8 2.9 78.4 3.4
11 33 73.9 3.1 57.7 2.6 16.2 3.0 78.2 3.5
12 33 74.7 3.7 57.9 2.4 16.8 3.3 77.6 3.5
13 31 75.8 3.7 57.9 2.4 17.9 3.4 76.5 3.6
14 30 77.0 3.6 58.4 2.5 18.6 3.2 75.9 3.4
15 26 78.0 3.9 59.1 2.4 18.9 3.5 75.9 3.5
16 27 78.2 4.0 59.0 2.2 19.1 3.3 75.6 3.3
17 25 77.9 3.9 58.7 2.7 19.2 3.1 75.4 3.3
18 22 79.1 4.1 59.1 2.4 19.9 3.7 75.0 3.8
(AG) Antegonion; (J) jugale.
Table 23-4 Mean cephalometric values (for selected ages) measured from radiographs (From Cortella et al.25)
(y) Mean SD Mean SD Mean SD Mean SD
6 71.6 3.0 56.3 2.3 15.3 2.4 78.6 2.8
9 77.1 3.4 60.6 2.6 16.5 3.0 78.7 3.2
12 81.0 3.9 62.7 2.5 18.3 3.6 77.6 3.5
15 85.0 4.3 64.5 2.6 20.6 3.8 75.9 3.5
18 86.4 4.5 64.7 2.7 21.8 4.1 74.9 3.8
rate of growth of maxillary width apparently slows down

Table 23-5 Percentage of radiographic enlargement before that of mandibular width. Growth in width of the
(%) relative to age and corresponding distance between jaws is reported to decline to a slower rate earlier than
film and porionic axis (FPD) (From Cortella et al.25) sagittal and vertical development, except in the posterior
Age (y) areas where the jaws grow wider as they grow in length
56
Females Males FPD (mm) Enlargement (%) posteriorly. The present findings support this observation,
5 6 90 5.9 as J-J and AG-AG are measurements in the posterior
regions of the jaws.
10 8 100 6.6
The greater growth observed in the mandible relative to
10.5 9.5 110 7.2
the maxilla suggests the presence of a compensatory
14 12 120 7.9 mechanism that allows the preservation of normal occlu-
> 18 15 130 8.5 sion (no crossbite) between the posterior teeth. Indicative
> 18 140 9.2 of such a mechanism is the increase in maxillary intermolar
55
width (computed from Moorrees’s studies ), which repre-
280
a b
Fig 23-12 (a) Maxillary width (J-J) in females and males as measured on radiographs and corrected for radiographic enlargement. (b) Mandibular width (AG-AG)
in females and males as measured on radiographs and corrected for radiographic enlargement. (Adapted from Cortella et al.25)
6,7
sents about 52% of the increase in J-J between ages 6 and The Rocky Mountain (RM) analysis, widely used for diag-
18 years, while the widening of the mandibular intermolar nosis of transverse relationships between the jaws (see Fig 23-
distance is about 17% that of AG-AG. Diagnosis of trans- 6), includes norms from ages 9 through 16 years. These norms
verse skeletal discrepancy versus dentoalveolar crossbite are greater than the corrected Bolton norms for both jaws;
ultimately depends on determining the range of normalcy however, the RM mandibular norms are smaller than the cor-
in the position and axial inclination of the posterior teeth responding radiographic Bolton values, and the RM maxillary
relative to the respective jaw and to each other. The prob- norms are larger than the Bolton norms. Consequently, the
lem in such determinations is the difficulty in tracing, and differences between the radiographic Bolton maxillary and
thus the reliability of reproducing the maxillary molars on mandibular widths are greater than the RM differences. This
PA cephalograms (see below). differential suggests that the Bolton norms allow for a smaller
maxillary width than the RM method. If the corrected Bolton
Inference on cephalometric analysis values should be used as norms, with one SD deemed with-
Radiographic enlargement is a critical factor in cephalo- in the range of normalcy, the average differences between
60
metrics. The adjusted norms (see Table 23-3) offer a Bolton and RM maxillary-mandibular relations (differences or
guideline for diagnosis, as the percent enlargement relative ratios) may not be clinically significant (Table 23-6). However,
to the film–porionic axis distance (FPD) (see Table 23-5) the diagnosis of maxillary and/or mandibular absolute width
may be applied in any clinical setting to correct for the may differ with the method of analysis.
standard FPD adopted in that setting, or for a specific indi- In Table 23-6, the Bolton and RM norms are compared
vidual FPD if the radiographic method involves placing the with other available data of Austrian children between the
12
film near the tip of the nose. Ghafari et al suggested the ages of 6 and 15 years.53 In these children, absolute (radi-
use of 13 cm as a practical standard FPD until a universal ographic uncorrected) measurements of maxillary and
standard is adopted. The corresponding enlargement fac- mandibular widths were greater than the Bolton and RM
tor is 8.5% (see Table 23-5). norms, but the computed ratios of maxillary and mandibu-
The enlargement factors vary at different ages and affect lar widths were closer to the Bolton ratios. Differences may
mandibular width more than maxillary width because AG- be related to methodology: besides a difference in the
AG is always larger than J-J. Thus, the differences between number of subjects, longitudinal records were used in the
AG-AG and J-J diverge from the differences calculated for Bolton study, whereas cross-sectional records apparently
the nonenlarged measurements. In contrast, the ratios of J- were used in the others. Furthermore, the film-object
J:AG-AG are not different between enlarged and corrected determination is variable in the Bolton population (see
measurements and should be considered a more accurate Table 23-5) and fixed in the Austrian sample (probably at 15
diagnostic guideline, particularly if the radiographic values cm, the European standard, a greater distance than in the
are not adjusted for enlargement. Bolton and most likely the RM studies), leading to greater
281
Table 23-6 Mean width of the maxilla and mandible, the difference between them, and their ratios as found in the Bolton,
Rocky Mountain, and Austrian samples
(y) Bolton* RM† Aus‡ Bolton* RM† Aus‡ Bolton* RM† Aus‡ Bolton* RM† Aus‡
6 71.6 78.5 56.3 61.0 15.3 17.5 78.6 77.8
9 77.1 76.0 82.7 60.6 62.0 63.8 16.5 14.0 18.9 78.7 81.6 77.2
12 81.0 82.0 85.4 62.7 63.8 64.6 18.3 16.4 20.8 77.6 77.8 75.8
15 85.0 84.4 91.2 64.5 65.6 67.4 20.6 18.8 23.8 75.9 77.7 74.0
18 86.4 85.8 64.7 66.2 21.8 19.6 74.9 77.2
*Bolton sample, from Cortella et al.25
†Rocky Mountain sample, after Grummons and Ricketts.6 (Ratios computed from available means.)
‡Austrian sample, from Athanasiou et al.53 (Differences and ratios computed from available means.)
distortion and measurements. The significant disparity the premise that deviations from normal PA relationships
between the three populations, presumably all Caucasian, must be corrected to a level comparable to normal rela-
underlines the need to establish more universal standards tionships and eventually normal adult size.
for PA cephalometry and to incorporate more proportional
and angular measurements in the frontal analysis. The control records included PA cephalograms of 30 sub-
PA cephalometry underscores the qualification of jects, at ages 10 and 18 years, from the Bolton-Brush Growth
cephalometric evaluation as only a guide to diagnosis. Center longitudinal database described above. The sub-
Even the most widely accepted measure of mandibular group used in this study included 16 females and 14 males
width (AG-AG) is subject to significant errors. Legrell37 sug- selected on assumptions regarding growth. The first group
gested that neither antegonion nor gonion can be used included subjects at age 10 years, a prepubertal age in boys
routinely as valid landmarks. The fact that cross-referencing (more than 2 SDs less than the average age of peak height
frontal landmarks with their counterparts on lateral velocity57) and in most girls (more than 1.5 SDs less than the
cephalograms improves their identification3,13,37 empha- average age of peak height velocity). Age 18 was selected for
sizes the importance of 3-D analysis and imaging on the the young adult group because it is a time when most
validity of identification and consequently the accuracy of growth has been completed in girls and in the majority of
diagnosis. boys.
The treatment group included the pretreatment PA
cephalograms of 24 patients (16 females and 8 males) treat-
ed with rapid maxillary expansion during the year of investi-
Orthopedic treatment with maxillary gation. Their average ages were 10.50 ± 0.89 years (females)
expansion and 10.01 ± 0.79 years (males). Patients were restricted to
around age 10 years for two reasons: their pretreatment
The aims and corresponding rationale of this investigation record could be compared with the 10-year-old norms, and
were to: maxillary expansion could be achieved without recourse to
surgical osteotomy, as might be necessary at postpubertal
1. Evaluate on PA cephalograms transverse craniofacial ages to achieve orthopedic split of the palate.61 The fixed
relationships and longitudinal changes on the basis of maxillary expander had acrylic palatal coverage (Haas type62)
new and available linear measurements and through and was activated twice per day (approximately 0.5 mm).
the introduction of angular measurements that are pre- All cephalometric distances were adjusted for radi-
sumed to be less variable than distances. The measure- ographic distortion by subtracting the percentage of
ments were performed in only two age groups repre- enlargement, which was computed on the basis of the dis-
senting pre- and postpubertal girls and boys. tance between porionic axis and film for both the Bolton-
2. Compare the generated normative data with the trans- Brush group and the treatment group.16 Distances present-
verse skeletal pattern of patients whose treatment included in this chapter are the corrected measures.
ed rapid maxillary expansion. Underlying this evaluation is
282
Table 23-7 Comparison of selected craniofacial distances (means and SDs in mm) among and between untreated (ages 10
and 18 years) and treated (about age 10 years) groups of males and females (From Huertas and Ghafari.16)*
CO-CO (SD) J-J (SD) AG-AG (SD) [(AG-AG)–(J-J)] (SD)
Males
Untreated (n = 14)
Age 10 years 53.16 (2.39) 58.64 (2.55) 73.43 (3.32) 14.79 (3.00)
Age 18 years 57.05 (3.10) 61.50 (2.49) 79.10 (4.04) 17.60 (3.41)
P value .001 .002 .0001 .001
Treated (n = 8) 52.78 (2.11) 54.79 (3.81) 75.73 (5.04) 20.94 (2.65)
P value NS .02 NS .0001
Females
Untreated (n = 16)
Age 10 years 54.71 (3.20) 57.57 (2.89) 73.08 (3.14) 15.52 (2.62)
Age 18 years 57.70 (3.39) 59.05 (2.65) 76.75 (2.82) 17.70 (3.15)
P value .0001 .007 .0001 .001
Treated (n = 16) 53.12 (3.01) 54.31 (2.81) 73.67 (3.63) 19.36 (3.46)
P value NS .003 NS .001
(CO) Center of orbit; (J) jugale; (AG) antegonion.
*Treated groups were compared with untreated groups at age 10 years.
All statistical comparisons were made with the t-test; P < .05.
Table 23-8 Comparison of distances (means and SDs in mm) between right and left molars among and between untreated
and treated groups (From Huertas and Ghafari.16)*
Maxillary distances (mm) Mandibular distances (mm)
Crowns Apices Crowns Apices
Males
Untreated (n = 14)
Age 10 years 51.22 (3.14) 47.22 (3.31) 46.91 (2.73) 54.94 (3.15)
Age 18 years 50.57 (2.71) 47.84 (3.70) 47.22 (2.58) 56.09 (2.97)
P value NS NS NS NS
Treated (n = 8)* 48.08 (2.95) 42.93 (2.72) 46.01 (3.51) 53.97 (3.13)
P value NS NS NS NS
Females
Untreated (n = 16)
Age 10 years 50.80 (3.00) 44.00 (3.67) 45.74 (3.20) 53.95 (3.49)
Age 18 years 49.52 (2.14) 44.13 (3.27) 44.65 (2.33) 52.96 (3.02)
P value .05 NS NS NS
Treated (n = 16) 47.95 (2.31) 39.68 (3.70) 44.03 (2.06) 53.58 (2.76)
P value .005 .002 NS NS
*Treated groups were compared with untreated groups at age 10 years.
Normative data The distance CO-CO, a surrogate measure of cranial

In both gender groups, statistically significant differences width, was almost equal to J-J in 18-year-old girls but high-
were found between ages 10 and 18 years for the distances ly correlated with AG-AG at age 18 years in both gender
CO-CO, J-J, AG-AG and for the difference between AG-AG groups (r = 0.71, P = .002 in males; r = 0.77, P < .0001 in
and J-J (see Fig 23-2; Table 23-7). Most statistically signifi- females). Low to moderate correlations were found
cant differences between gender groups occurred at age between CO-CO and J-J, between J-J and AG-AG, and
18 years and involved the distances J-J (P = .009), maxillary between these and their differences (the highest correla-
(P = .004) and mandibular (P = .005) interapical distances, tion was r = –0.63, P = .005, between 10- to 18-year incre-
and maxillary intermolar distance U6-6C (P = .005). ments of J-J and the difference [AG-AG]–[J-J]).
283
Table 23-9 Comparison of selected angular measurements (means and SDs in degrees) among and between untreated (ages
10 and 18 years) and treated (about age 10 years) groups (From Huertas and Ghafari.16)*
J-Cr-AG (degrees) J-CO-AG (degrees)
Right (SD) Left (SD) Average (SD) Right (SD) Left (SD) Average (SD)
Males
Untreated (n = 14)
Age 10 years 9.19 (1.88) 8.37 (2.77) 8.77 (2.32) 3.25 (1.67) 4.48 (2.78) 3.86 (2.22)
Age 18 years 8.00 (2.90) 9.53 (3.21) 8.76 (3.05) 5.23 (2.10) 4.12 (2.39) 4.68 (2.24)
P value NS NS NS NS NS NS
Treated (n = 8)* 3.64 (2.29) 4.22 (1.10) 8.86 (1.63) 7.66 (1.66)
P value .0001 .0001 .0001 .003
Females
Untreated (n = 16)
Age 10 years 7.72 (2.09) 8.47 (2.13) 8.09 (2.51) 5.09 (1.97) 5.32 (2.63) 5.20 (2.30)
Age 18 years 8.47 (2.02) 8.63 (1.73) 8.55 (1.86) 5.55 (2.00) 5.96 (2.58) 5.75 (2.28)
P value NS NS NS NS NS NS
Treated* 4.94 (3.23) 4.43 (3.40) 7.01 (3.08) 8.42 (2.82)
P value .007 .0001 .05 .003
(J) Jugale; (Cr) crista galli; (AG) antegonion; (CO) center of orbit; (NS) not significant.
*Treated groups were compared with untreated groups at age 10 years. All comparisons were P < .05.
Table 23-10 Comparison of selected angular measure-

ments (means and SDs in degrees) in untreated groups at 1
ages 10 and 18 years (From Huertas and Ghafari.16) 1
J-Cr-midline AG-Cr-midline Differences
Age 10 years 26.02 (1.91) 35.03 (2.69) 9.01 (2.35)
2
Age 18 years 23.43 (1.73) 32.42 (2.39) 9.02 (2.88)
3 3
P value* NS NS NS
*T-test; P < .05. 4 4
5
Fig 23-13 Basic areas of differentiation in craniofacial development

between adolescent males (m) and females (f): 1-Supraorbital ridges—(m)
well developed, (f) virtually absent; 2-zygomatic prominences (cheek-
bones)—(m) large, (f) small; 3-mastoid processes—(m) large, (f) small and
delicate; 4-gonial angle—(m) prominent lipping, (f) rounded; 5-mandibular
symphysis—(m) prominent, (f) rounded. (From Broadbent et al.3)
In both gender groups, the increase in mandibular maxillary and mandibular first molars correlated at levels
width (5.5 mm in males; 3.9 mm in females) was more than greater than r = 0.7 in all age and gender groups (0.86 < r <
twice that of maxillary width (2.4 mm in males; 1.2 mm in 0.90; P = .0001) except 10-year-old girls (r = 0.63; P = .003).
females). The ratio J-J:AG-AG was slightly greater in boys New measures of jaw relations included the right and
(80.3%) than girls (78.8%) at age 10 years and at age 18 left angles between jugale, antegonion, and either crista
years (77.7% in males; 76.9% in females). galli in the midline or the center of the orbit on the corre-
Dentoalveolar measurements, represented by the dis- sponding lateral side (see Fig 23-2; Tables 23-9 and 23-10).
tances between right and left first molars at the level of the J-Cr-AG and J-CO-AG were highly correlated with the linear
crowns and apices, were similar at both age groups in both difference between J-J and AG-AG at both age groups (0.64
genders (Table 23-8). Distances between the crowns of < r < 0.85; .0001 < P < .01) in males; only J-CO-AG exhibited
284
Fig 23-14 Relation between maxillary width (J-J)

and mandibular width (AG-AG) in 10-year-old
untreated boys (in black) and corresponding group
(in blue) treated with rapid maxillary expansion
(RME). Differences between the slopes were not sta-
tistically significant. The majority of treated boys (7
of 8) had a smaller than average maxillary width and
6 of 8 had a larger than or equal to average
mandibular width. (From Huertas and Ghafari.16)
Fig 23-15 Relation between maxillary width (J-J)

and mandibular width (AG-AG) in 10-year-old
untreated girls (in black) and corresponding treated
group (in blue) treated with rapid maxillary expan-
sion (RME). Differences between the slopes were not
statistically significant. The majority of treated girls
(13 of 16) had a smaller than average maxillary
width, and 10 of 16 had a larger than or equal to
average mandibular width. (From Huertas and Gha-
fari.16)
such correlation in females (0.66 < r < 0.84; .0001 < P < .003). The results support previous conclusions that different
Thus, the angles J-CO-AG exhibited higher correlations normative data should be used for males and females when
than J-Cr-AG with the linear difference ([AG-AG]–[J-J]) linear measurements are considered.4,6–11,25,43,63,64 Sexual
between the jaws. In addition to determining asymmetry dimorphism in craniofacial development has been
between right and left sides of jaw, the angles J-Cr-midline described by Broadbent et al3 in specific areas of the head
and AG-Cr-midline (see Table 23-10), or the corresponding (Fig 23-13). The newly introduced angular measurements,
measures relating J and AG to the vertical through CO par- like angular measurements in the sagittal plane, are similar in
allel to the midline, help determine which of the jaws devi- both genders and can be used for both (see Table 23-9).
ates from the norm.
285
expected increase in mandibular width with growth. In this

Table 23-11 Incidence and types of malocclusion in context, it would be important to investigate gender differ-
treated group (From Huertas and Ghafari.16)
ences in treatment needs, because the ultimate difference in
Posterior Class II Class III Other maxillary width (and in maxillary and mandibular discrepancy)
crossbite (%) (%) (%) (%) between pretreatment and normal values is greater in males
Males (n = 8) 4* (50) 5* (62.5) 0 1 (12.5) (see Table 23-7, Figs 23-14 and 23-15). In contrast, the corre-
Females (n = 16) 7 (44) 6 (37.5) 1 (6) 2 (12.5) sponding differences in the intermolar (crown) distances are
Total (n = 24) 11 (46) 11 (46) 1 (4) 3 (12.5) rather similar in both genders (see Table 23-8).
*Combined Class II and posterior crossbite: n = 2 (25%).
Application of PA
Maxillary expansion Cephalometry
In the treated group, maxillary skeletal and dentoalveolar
widths were narrower (.003 < P < .02) than in the corre-
sponding Bolton-Brush normative group (see Tables 23-7 Diagnosis
to 23-9), and the difference between maxillary and
mandibular widths was greater. Linear regressions of the Unlike the lateral headfilm, which provides information on
relations between J-J and AG-AG show almost parallel sagittal and vertical relationships among and between the
slopes for control and treated groups in both genders but jaws and teeth, data from the frontal radiograph are typically
at a lower level consistent with smaller J-J distances in the focused on assessment of asymmetry and widths of the jaws
treated group (Figs 23-14 and 23-15). and cranium. Quantitative measures in the vertical plane are
The variances in control and treated groups were similar more prone to errors with the PA cephalogram related to
in girls and boys and the differences between the slopes of head tilt than those derived from the lateral headfilm. Also,
maxillary and mandibular regressions were not statistically the skew to sagittal cephalometrics reflects the sustained use
significant, a finding illustrated by the nearly parallel regres- of Edward H. Angle’s sagittally defined classification of maloc-
sion lines (see Figs 23-14 and 23-15). This result would sug- clusion, which provides a universal guide to diagnosis and
gest that an increased maxillary width (J-J) would normal- treatment planning.1
ize the maxillomandibular relationship of the treated group Consequently, the frontal cephalogram is still not used
to approach control values in the treated group. routinely in clinical orthodontics and represents a minor
The rationale of the treating orthodontists for planning percentage of total cephalometric studies in the literature.
maxillary distraction was discounted as an inclusion criteri- Yet, critical information has been obtained that completes
on. The results revealed that a majority of the children had the 3-D picture of growth, diagnosis, and treatment. In
posterior crossbites (46%) and/or Class II skeletal relations addition to the growth data presented earlier, information
(46%) (see Figs 23-14 and 23-15; Table 23-11). Maxillary arch can be drawn from the PA record about the nasal cavity (no
form is known to be narrower in Class II malocclusions.65,66 gender difference in nasal width,67,68 turbinate hypertro-
Several female patients who had a close to normal relation phy69), canine impaction,70 cervical lordosis,71 and key find-
between J-J and AG-AG had posterior crossbites (see Fig ings in patients with cleft palate39,45,46,72 and other craniofa-
23-15), suggesting that these malocclusions were of a den- cial anomalies.73
toalveolar rather than skeletal nature. Expansion apparent-
ly was planned in some children for space creation or
esthetic considerations to enhance facial appearance dur- Transverse growth
ing smile. A narrow maxillary arch influences the width and
configuration of the space between the maxillary lateral Previous studies16,25 indicate that transverse development
teeth and the corner of the lips during smile.24 When of the jaws is characterized by differential growth between
enlarged, this space is known as a black space or corridor. maxilla and mandible. Mandibular width proceeds, on
Stability of the occlusion is related to the width of the den- average, at a ratio of 2:1 relative to maxillary width
tal arches and by extension the underlying jaws. Follow-up between ages 10 and 18 years. This conclusion is best illus-
studies are warranted to determine short- and long-term trated in the composite of the annual tracings of the
effects of maxillary expansion, not only to gauge the stability frontal Bolton standards,3 which also show the uniformity
of the results, but also to evaluate whether the widened max- of morphologic patterns from 3 to 18 years (Fig 23-16).
illa is closer to adult norms and whether the maxilla should be In the interval of 10 to 18 years, a differential ratio has
overexpanded to adult proportions in anticipation of the also been described in the vertical and sagittal
286
Application of PA Cephalometry
tions at both ages. This finding might conceptually support

the functional matrix premise of functional requirements influ-
encing optimal form.79 A posterior crossbite may result from
insufficient dentoalveolar compensations. On the other hand,
mandibular crowding may occur from excessive compensa-
tory lingual inclination of the mandibular buccal teeth. A
potential therapeutic corollary may be attached to these den-
toalveolar compensations. The clinician, particularly in the
presence of severe skeletal discrepancies, should control over-
compensation by normalizing the inclination of the posterior
teeth, usually through maxillary expansion and corresponding
uprighting of mandibular posterior teeth. Excessive compen-
sations also may affect periodontal health, including gingival
recession and buccal bone loss. Therefore, early treatment
may be needed to correct developing transverse problems.
Treatment
When assessing the relevance of transverse norms to
Fig 23-16 Frontal annual Bolton standards from 3 to 18 years. Note the transverse orthopedics, it may be argued that whether
greater increase of mandibular width than maxillary width during facial the decision to treat is related to the posterior crossbite,
growth. (From Broadbent et al.3) space management (creation), or esthetic consideration,
clinical impression anticipates or foregoes cephalometric
findings. Moreover, the target of correction tends to be
planes.54,74–77 Correlations between the differential maxillo- the maxilla even if the mandible is the discrepant jaw,
mandibular changes in all planes of space are not known; it because maxillary expansion is easier and more feasible
is also not known how their timely interaction contributes than restraining the transverse growth or constricting the
to the development of malocclusion. Consequently, early mandible. Even if surgery is involved, osteotomy of the
intervention to correct a developing malocclusion would maxilla is a less morbid procedure than surgical narrowing
depend not only on intercepting unfavorable discrepan- of the mandible.
cies in differential ratios between the jaws, but also (and The fact that the majority of the treated children in the
just as importantly) on when this treatment is rendered. study by Huertas and Ghafari16 had narrow maxillary width
Changes observed in the posterior width of the maxilla (see Figs 23-14 and 23-15) seemingly supports discarding
(J-J) and mandible (AG-AG) are consistent with observa- the PA record, given the prevalence of maxillary correction.
tions by Björk and Skieller,59 who measured growth in max- Although nearly all available analyses contain information
illary width between posterior implants in nine boys (10 to about the individual position of the maxilla and the
11 years to adult age), and by Baumrind and Korn,78 who mandible, the information is often irrelevant to treatment of
evaluated the lateral displacement of metallic implants in transverse malocclusion. However, the PA cephalogram, like
the mandibles of 31 subjects (8.5 to 15.5 years). In addition, sagittal cephalometry, is only a guide to assist proper diag-
the authors of both studies reported that posterior width nosis. Both records complete the 3-D evaluation of the
grows more than the anterior breadth of the jaws. This find- patient and support the rationale for treatment, not to men-
ing may account for Grayson et al’s observation of more tion their undeniable value in research. Unfortunately, the
severe asymmetry in the posterior than the anterior region overlap of structures on the PA film renders the identifica-
of the head (see Fig 23-11).42 tion of molars, and consequently the diagnosis of posterior
While the maxillomandibular growth differential is 2:1 alveolar inclination, difficult. To lessen error, Huertas and
between ages 10 and 18 years, posterior teeth and associated Ghafari16 introduced identification of the molar teeth as the
alveolar bone compensate for this discrepancy. In this time line connecting the most buccal points on crowns and
interval, normal transverse occlusion is maintained (as per roots at the level of the apices. The new CT scan technology
inclusion criteria), dentoalveolar width at the level of first (see chapter 21), applied to a patient’s head positioned in a
molars (between right and left buccal surfaces of crowns and cephalostat, therefore yielding more reproducible and reli-
apices) seems to be stable (see Table 23-8), and maxillary and able images, will also facilitate recognition of the landmarks
mandibular intermolar (crown) distances exhibit high correla- and structures that are currently difficult to identify.
287
Bizygomatic width
LM
Fig 23-18 Graphic representation of the law of Izard. The ratio of maxillary
Tr
arch maximal width (LM) to greatest facial width (bizygomatic distance 2 to
Eu Eu 3 cm frontal to external auditory meati) is approximately 1:2. (Skull drawing
Ft Ft adapted from Faigin.23)
N
T T
Zy Zy
Go Go
b c GN
Fig 23-17 (a) Anthropometric measures are computed from Farkas’s data43 in • Width of the head, the distance between right and left
comparative linear projections over age. (Tr-Gn) trichion-gnathion height; eurion (Eu-Eu), the most prominent lateral point on each
(Eu-Eu) Bi-eurion head width; (T-T) bitragion skull base width; (Zy-Zy) bizy- side of the skull in the area of the parietal and temporal
gion face width; (Ft-Ft) forehead width; (Go-Go) bigonial mandibular width; bones
(N-Gn) nasion-gnathion height. (b,c) Nasion-gnathion height nearly equals • Width of the forehead (Ft-Ft), measured between the
the forehead width. (Drawings adapted from Farkas.43) points located laterally from the temporal lines
• Width of the skull base, or bitragion (T-T) diameter
• Facial width, between the zygions (Zy-Zy), also known as
bizygion diameter, upper facial width, or maximum
interzygomatic breadth
• Width of the mandible, the distance between the
gonions (Go-Go), also termed bigonial diameter or lower
Anthropometric perspective facial width; measured with calipers firmly pressed
against the bony surfaces because of the varying thick-
ness of the soft tissue covering the mandibular angles
Cephalometry does not replace but complements anthro-
pometry, because facial esthetics cannot be evaluated According to Farkas,43 in the midline the physiognomic
through analysis of hard tissue (bone and teeth) only. While height of the face is defined between trichion and gnathion
analysis of facial proportions is not detailed in this chapter, and the morphologic height between nasion and gnathion.
attention is drawn to anthropometric measurements that Comparison of these distances reveals that the nasion-
correspond to the cephalometric characteristics discussed gnathion height is almost equal to the forehead width (see
above. Critical widths defined by Farkas43 (see Fig 23-17) graph in Fig 23-17), describing a square in which the face fits,
include: at least in the average facial pattern (mesoprosopic).
288
Conclusions
Attempts have been made to relate the maxillary arch Important premises underlie the transition to 3-D
width to facial width. Izard defined an index of greatest cephalometry, which seemingly resembles the early days
maxillary arch width (LM, at the level of the second or first of cephalometrics, with the benefit of decades of cephalo-
molars, whichever is the largest) to bizygomatic width metric knowledge to help guide future applications:
(BZO, measured as the widest facial width, between 2 and
3 cm in front of the external auditory meati, less 10 mm of 1. Testing is required of the existing linear, angular, and
estimated skin thickness) in a LM:BZO ratio of nearly 1:2 proportional norms, because the measurements are
(Fig 23-18).80 But the relationship has been discredited by made on anatomic landmarks digitized in 3-D coordi-
many authors because the ratio did not hold in controlled nates, and not on their bidimensional projection.
studies, particularly in euryprosopic facial types80 (see Fig 2 Until norms are confirmed, or new ones developed for
23-9). The relation between facial and dental indices is 3-D images, existing linear and angular measurements
probably a factor of age. The demonstrated increase of could be performed on planar images because 3-D data
mandibular width (AG-AG) at twice the amount of maxil- can be rendered as two-dimensional (2-D) projection
lary width (J-J) between the ages of 10 and 18 years is bal- similar to a radiograph. Research should determine
anced by tipping of the maxillary molars to maintain a whether bidimensional analyses will still be used, albeit
proper buccolingual relation.16 Confounding the issue of supplementing 3-D application.
diagnosis and treatment planning is the width of the com- 3. Significant issues and findings based on traditional
missure upon smiling, whereby an increased space cephalometry and related to diagnosis, growth, and
between the maxillary lateral teeth and the corners of the treatment must be revisited for substantiation or clarifi-
lips during smile creates a black space that detracts from cation with a 3-D tool that reproduces the head more
optimal esthetics.24 realistically and discloses details that are not as accurate-
ly definable with the “old” technology. Some authors82,84
have already tested applications on images derived from
the medical CT scanner, or from the more cephalomet-
Transition to 3-D cephalometry rically suitable cone-beam scanners in which the head is
held in a more stable position. However, further techno-
logical standardization should be expected before more
3-D craniofacial imaging (see chapter 21) is expected to encompassing research can proceed.
replace many conventional radiographic (and even non-
radiographic) orthodontic records.81,82 These advantages
must be noted:
1. The image to head size is displayed in a ratio of 1:1.

Conclusions
2. Landmarks and references are better defined than on
traditional cephalometric radiographs, and their identifi- 1. Transverse evaluation of the craniofacial skeleton is criti-
cation should be more accurate with further technolog- cal to the complete cephalometric diagnosis. Newly
ical developments. introduced linear and angular measurements yield
3. New landmarks and analyses might be introduced, and advantages to the PA craniofacial cephalometric record
geometric cephalometric landmarks (eg, articulare) not provided by earlier analyses. Angular norms can be
might be eliminated or would need verification for sus- used similarly for both males and females.
tained use. 2. A great number of anatomic structures and landmarks
4. Norms may combine the advantages of cephalometrics are not readily identifiable on PA films and are subject to
and anthropometrics. error; proportionate analysis tends to reduce the impact
5. 3-D records may not have to be used in the old cephalo- of such errors.
metric ways and their use shall benefit from advances in 3. New technologies should incorporate all dimensions
related fields (geometric morphometrics).82 into one computerized record that will improve struc-
6. Since more accurate insights into anatomy, diagnosis, ture identification and obviate the need to cross-refer-
position of teeth, response to treatment should be forth- ence landmarks and structures from the lateral and PA
coming,83 3-D frontal cephalometrics should improve radiographs.
understanding of orthodontic mechanics and their 4. Research on methodological errors indicates that the PA
effect on the dentition, particularly the posterior teeth, cephalogram is subject to operational error related to
because their imaging has been deficient with conven- asymmetry of ears and head position, thus requiring
tional cephalography. proper evaluation of the head before and after it is posi-
289
tioned in the cephalostat and further standardization of 4. Grummons DC, Kappeyne van de Coppelo MA. A frontal asym-
the record. metry analysis. J Clin Orthod 1987;21:448–465.
5. Studies of validity of reliability of the PA cephalogram 5. Moorrees CFA. Orthodontics and dentofacial orthopedics—Past,
present and future, part 1. Kieferorthop 1998;12:17–26.
and corresponding reference lines demonstrate that in
6. Grummons D, Ricketts RM. Frontal cephalometrics: Practical
the study of asymmetry, vertical reference lines connect- applications, part 2. World J Orthod 2004;5:99–119.
ing midline points to the anterior nasal spine are less 7. Ricketts R. Perspectives in the clinical application of cephalo-
accurate than perpendiculars to horizontal lines con- metrics: The first fifty years. Angle Orthod 1981;51:115–150.
necting bilateral cranial landmarks (particularly fronto- 8. Ricketts RM, Roth RH, Chaconas SJ, Schulhof RJ, Engel GA. Ortho-
zygomatic suture and orbital landmarks). dontic Diagnosis and Planning. Denver: Rocky Mountain Data
6. In addition to visualizing asymmetry of structures, the PA Systems, 1982.
9. Sassouni V. The Face in Five Dimensions. Philadelphia: Growth
record’s practical applications have been limited to the
Center Publication, 1955.
relationship between maxillary and mandibular widths. 10. Bergman R. Practical application of the PA cephalometric head-
In patients with posterior crossbite, available research film. Orthod Rev 1988;2:20–26.
revealed a tendency for reduced maxillary width. Similar 11. Athanasiou AE, Van der Meij AJW. Posteroanterior (frontal)
to the evaluation of sagittal problems, clinical impres- cephalometry. In: Athanasiou AE (ed). Orthodontic Cephalome-
sion apparently anticipates cephalometric findings. try. London: Mosby-Wolfe, 1995:141–161.
7. Utilization of frontal cephalometry has been more elec- 12. Ghafari J, Cater PE, Shofer FS. Effect of film-object distance on
posteroanterior cephalometric measurements: Suggestions for
tive than sagittal cephalometry because the target of
standardized cephalometric methods. Am J Orthod Dentofacial
correction tends to be the maxilla (maxillary expansion), Orthop 1995;108:30–37.
even if mandibular discrepancy exists, because manipu- 13. Hsiao TH, Chang HP, Liu KM. A method of magnification correc-
lation of the mandible through constriction or expan- tion for posteroanterior radiographic cephalometry. Angle
sion of the bone is difficult and not recommended. If Orthod 1997;67:137–142.
surgical intervention is necessary, osteotomy of the 14. Thurow RC. Otic axis locator: Closing the accuracy gap in
maxilla is a less morbid procedure than surgical narrow- cephalometrics and cast mounting. Am J Orthod Dentofacial
ing or expansion of the mandible. Nevertheless, even Orthop 2000;117:298–302.
15. Solow B. The pattern of craniofacial associations: A morphologi-
the diagnosis of maxillary and mandibular width rela-
cal and methodological correlation and factor analysis study on
tionships warrants the incorporation of the PA radi- young adult males. Acta Odontol Scand 1966;suppl 46.
ograph into comprehensive orthodontic and dentofa- 16. Huertas D, Ghafari J. New posteroanterior cephalometric norms:
cial diagnosis. Comparison with craniofacial measures of children treated with
8. Cross-referencing anthropometric and cephalometric palatal expansion. Angle Orthod 2001;71:285–292.
measures helps improve a comprehensive diagnosis, 17. Martin R, Saller K. Lehrbuch der Anthropologie, vol 1, ed 3.
but much research is still required when soft tissues in Stuttgart: Gustav Fischer Verlag, 1957:625–643.
18. Moyers RE. Handbook of Orthodontics, ed 4. Chicago: Year Book
action are considered (during smile or speech).
Medical, 1988.
9. 3-D cephalometrics may reveal advantages heretofore 19. Trpkova B, Prasad NG, Lam EWN, Raboud D, Glover KE, Major PW.
unrecognized because of the constraints of 2-D frontal Assessment of facial asymmetries from posteroanterior cephalo-
analysis. These benefits will encompass diagnosis and grams: Validity of reference lines. Am J Orthod Dentofacial
evaluation of growth and treatment, particularly the Orthop 2003;123:512–520.
effect of orthodontic mechanics on the dentition. 20. Schmid W, Mongini F, Felisio A. A computer-based assessment of
structural and displacement asymmetries of the mandible. Am J
Orthod Dentofacial Orthop 1991;100:19–34.
21. Wei SHY. Cranial width dimensions. Angle Orthod 1970;40:
141–147.
References 22. Betts NJ, Lisenby WC. Normal adult transverse jaw values
obtained using standardized posteroanterior cephalometrics
[abstract 1567]. J Dent Res 1994;73:298.
1. Moorrees CFA, Kalpins RI, Ghafari JG. Proportional analysis of the 23. Faigin G. The Artist’s Complete Guide to Facial Expression. New
human face in a mesh coordinate system. In: Jacobson A (ed). York: Watson-Guptill Publications, 1990.
Radiographic Cephalometry: From Basics to Videoimaging. 24. Ghafari J. Emerging paradigms in orthodontics—An essay. Am J
Chicago: Quintessence, 1995:197–215. Orthod Dentofacial Orthop 1997;111:573–580.
2. Gottlieb EL, Nelson AH, Vogels DS. JCO study of orthodontic 25. Cortella S, Shofer FS, Ghafari J. Transverse development of the
diagnosis and treatment procedures. Part 1, Results and trends. jaws: Norms for the posteroanterior cephalometric analysis. Am
J Clin Orthod 1990;25:145–156. J Orthod Dentofacial Orthop 1997;112:519–522.
3. Broadbent BH Sr, Broadbent BH Jr, Golden WH. Bolton Standards 26. Chidiac JJ, Shofer FS, Al-Kutoubi A, Laster LL, Ghafari J. Compari-
of Dentofacial Developmental Growth. St Louis: Mosby, 1975. son of CT scanograms and cephalometric radiographs in cranio-
facial imaging. Orthod Craniofac Res 2002;5:104–113.
290
References
27. Perillo MA, Shofer FS, Beideman RW, et al. Effect of landmark 49. Vogel CJ. Correction of frontal dimensions from head x-rays.
identification on cephalometric measurements. Clin Orthod Res Angle Orthod 1967;37:1–8.
2000;3:29–36. 50. Adams CP. The measurement of bizygomatic width on cephalo-
28. Ahlqvist J, Eliasson S, Welander U. The effect of projective errors metric x-ray films. Dent Pract 1963;14:58–63.
on cephalometric length measurements. Eur J Orthod 1986; 51. Baumrind S, Frantz R. The reliability of head film measurements.
8:141–148. 1—Landmark identification. Am J Orthod 1971;60:111–127.
29. El-Mangoury EH, Shaheen SI, Mostafa YA. Landmark identifica- 52. Ghafari J, Jacobsson-Hunt U, Higgins-Barber K, Beideman RW,
tion in computerized posterior-anterior cephalometrics. Am J Shofer FS, Laster LL. Identification of condylar anatomy affects
Orthod Dentofacial Orthop 1987;91:57–61. the evaluation of mandibular growth. Guidelines for accurate
30. Major PW, Johnson DE, Hesse KL, Glover KE. Landmark identifica- reporting and research. Am J Orthod Dentofacial Orthop
tion error in posterior anterior cephalometrics. Angle Orthod 1996;107:645–652.
1994;64:447–454. 53. Athanasiou AE, Droschl H, Bosch C. Data and pattern of trans-
31. Major PW, Johnson DD, Hesse KL, Glover KE. Effect of head orien- verse dentofacial structure of 6- to 15-year-old children: A pos-
tation on posterior anterior cephalometric landmark identifica- teroanterior cephalometric study. Am J Orthod Dentofacial
tion. Angle Orthod 1996;66:51–60. Orthop 1992;101:465–471.
32. Yoon YJ, Kim DH, Yu PS, Kim HJ, Choi EH, Kim KW. Effect of head 54. Bambha JK. Longitudinal cephalometric roentgenographic
rotation on anteroposterior cephalometric radiographs. Angle study of face and cranium in relation to body height. J Am Dent
Orthod 2002;72:36–42. Assoc 1961;63:776–799.
33. Houston WJB. The analysis of errors in orthodontic measure- 55. Moorrees CFA. The size of the dental arch. In: Moorrees CFA (ed).
ments. Am J Orthod 1983;83:382–390. The Dentition of the Growing Child. Cambridge, MA: Harvard
34. Prittiniemi P, Miettinen J, Kantomaa T. Combined effects of errors in Press, 1959:87–110.
frontal-view asymmetry diagnosis. Eur J Orthod 1996;18:629–636. 56. Proffit WR. Contemporary Orthodontics, ed 2. St Louis: Mosby,
35. Sharad M. An assessment of asymmetry in the normal craniofa- 1993:87–104.
cial complex. Angle Orthod 1978;48:141–148. 57. Tanner JM, Davies P. Clinical longitudinal standards for height
36. Zepa I, Huggare J. Reference structures for assessment of frontal and height velocity for North American children. J Pediatr
head posture. Eur J Orthod 1998;20:694–699. 1985;107:317–329.
37. Legrell PE, Nyquist H, Isberg A. Validity of identification of gonion 58. Hamill PVV, Drizd TA, Johnson CL, Reed RB, Roche AF, Moore WM.
and antegonion in frontal cephalometrics. Angle Orthod Physical growth: National Center for Health Statistics percentiles.
2000;70:157–64. Am J Clin Nutr 1979;32:607–629.
38. Broadbent BH. A new x-ray technique and its application to 59. Björk A, Skieller V. Growth of the maxilla in three dimensions as
orthodontia. Angle Orthod 1931;1:45–66. revealed radiographically by the implant method. Br J Orthod
39. Ishiguro K, Krogma WM, Mazaheri M. A longitudinal study of 1977;4:53–64.
morphological craniofacial pattern via PA x-ray headfilms in cleft 60. Dibbets JMH. Applicability of cephalometric standards: An
patients from birth to six years. Cleft Palate J 1976;13:104–126. appraisal of atlases. In: Trotman CA, McNamara JA Jr (eds). Ortho-
40. Van der Linden FPGM, Boersma H. Diagnosis and Treatment dontic Treatment: Outcome and Effectiveness, vol 30, Craniofa-
Planning in Dentofacial Orthopedics. London: Quintessence, cial Growth Series. Ann Arbor, MI: Center for Human Growth and
1987:81–86. Development, 1995:297–317.
41. Miyashita K. Contemporary Cephalometric Radiography. Tokyo: 61. Betts NJ, Vanarsdall RL, Barber HD, Higgins-Barber K, Fonseca RJ.
Quintessence, 1996. Diagnosis and treatment of transverse maxillary deficiency. Int J
42. Grayson BH, McCarthy JG, Bookstein F. Analysis of craniofacial asym- Adult Orthodon Orthognath Surg 1995;10:75–96.
metry by multiplane cephalometry. Am J Orthod 1983;84:217–224. 62. Haas AJ. Palatal expansion: Just the beginning of dentofacial
43. Farkas LG. Anthropometry of the Head and Face, ed 2. New York: orthopedics. Am J Orthod 1970;57:219–255.
Raven Press, 1994. 63. Christie TE. Cephalometric patterns of adults with normal occlu-
44. Moorrees CFA, Kean MR. Natural head position: A basic consid- sions. Angle Orthod 1977;47:128–133.
eration in the interpretation of cephalometric radiographs. Am J 64. Yavuz I, Ikbal A, Baydas B, Ceylan I. Longitudinal posteroanterior
Phys Anthrop 1958;16:213–234. changes in transverse and vertical craniofacial structures
45. Laspos CP, Kyrkanides S, Tallents RH, Moss ME, Subtelny JD. between 10 and 14 years of age. Angle Orthod 2004;74:
Mandibular and maxillary asymmetry in individuals with unilat- 624–629.
eral cleft lip and palate. Cleft Palate Craniofac J 1997;34:232–239. 65. Moorrees CFA, Grøn AM, Lebret LM, Yen PK, Frohlich FJ. Growth
46. Trotman CA, Papillon F, Ross RB, McNamara JA Jr, Johnston LE Jr. studies of the dentition: A review. Am J Orthod 1969;55:
A retrospective comparison of frontal facial dimensions in alve- 600–616.
olar-bone-grafted and nongrafted unilateral cleft lip and palate 66. Varrela J. Early developmental traits in Class II malocclusion. Acta
patients. Angle Orthod 1997;67:389–394. Odontol Scand 1998;56:375–377.
47. Molsted K, Dahl E. Asymmetry of the maxilla in children with 67. Snodell SF, Nanda RS, Currier GF. A longitudinal cephalometric
complete unilateral cleft lip and palate. Cleft Palate J 1990; study of transverse and vertical craniofacial growth. Am J
27:184–192. Orthod Dentofacial Orthop 1993;104:471–483.
48. Potter JW, Meredith HV. A comparison of two methods of 68. Haralabakis NB, Yiagtzis SC, Toutountzakis NM. Cephalometric
obtaining biparietal and bigonial measurements. J Dent Res characteristics of open bite in adults: A 3-D cephalometric eval-
1948;27:459–66. uation. Int J Adult Orthodon Orthognath Surg 1994;9:223–231.
291
69. Ghafari J. Therapeutic and developmental maxillary orthopedics: 76. Nanda RS. The rates of growth of several facial components
Evaluation of effects and limitations. In: Davidovitch Z, Mah J (eds). measured from serial cephalometric roentgenograms. Am J
Biological Mechanisms of Tooth Eruption, Resorption, and Orthod 1955;41:658–673.
Replacement by Implants. Boston: Harvard Society for the 77. Enlow DH. Facial Growth, ed 3. Philadelphia: Saunders, 1990:
Advancement of Orthodontics, 2004:167–181. 240–242.
70. Sambataro S, Baccetti T, Franchi L, Antonini F. Early predictive 78. Baumrind S, Korn EL. Postnatal width changes in the internal
variables for upper canine impaction as derived from posteroan- structures of the human mandible: A longitudinal three-dimen-
terior cephalograms. Angle Orthod 2005;75:28–34. sional cephalometric study using implants. Eur J Orthod
71. D’Attilio M, Epifania E, Ciuffolo F, et al. Cervical lordosis angle 1992;14:417–426.
measured on lateral cephalograms—Findings in skeletal Class II 79. Moss ML. The primary role of functional matrices in facial
female subjects with and without TMD: A cross-sectional study. growth. Am J Orthod 1969;55:566–577.
Cranio 2004; 22:27-44. 80. Chateau M. Orthopédie dentofaciale. Paris: Ed J Prelat, 1975:
72. Athanasiou AE, Hack B, Enemark H, Sindet-Pedersen S. Trans- 63–64.
verse dentofacial structure of young men who have undergone 81. Nakajima A, Sameshima GT, Arai Y, Homme Y, Shimizu N,
surgical correction of unilateral cleft lip and palate: A posteroan- Dougherty H Sr. Two- and three-dimensional orthodontic imag-
terior cephalometric study. Int J Adult Orthodon Orthognath ing using limited cone beam-computed tomography. Angle
Surg 1996;11:19–28. Orthod 2004;75:895–903.
73. Polley JW, Figueroa AA, Liou EJ, Cohen M. Longitudinal analysis 82. Halazonetis DJ. From 2-dimensional cephalograms to 3-dimen-
of mandibular asymmetry in hemifacial microsomia. Plast sional computed tomography scans. Am J Orthod Dentofacial
Reconstr Surg 1997;99:328–339. Orthop 2005;127:627–637.
74. Baughan B, Demirjian A, Lesveque GY, Lapalme-Chaput L. The 83. Walker L, Enciso R, Mah J. Three-dimensional localization of max-
pattern of facial growth before and during puberty as shown by illary canines with cone-beam computed tomography. Am J
French-Canadian girls. Ann Human Biol 1979;6:59–76. Orthod Dentofacial Orthop 2005;128:418–423.
75. Hunter CJ. The correlation of facial growth with body height and 84. Treil J, Casteigt J, Faure J, Madrid C, Borianne P, Jaeger M. Archi-
skeletal maturation at adolescence. Angle Orthod 1966;36: tecture cranio-facio-maxillo-dentaire. Un modèle tridimension-
44–54. nel. Applications en clinique orthodontique et chirurgie orthog-
natique. In: Encyclopédie médico-chirurgicale. Odontologie et
Stomatologie. Paris: Elsevier, 2000:23-455-E-40.
292

Posteroanterior Cephalometry: Craniofacial Frontal Analysis: Joseph G. Ghafari

Uploaded by

Copyright:

Available Formats

Posteroanterior Cephalometry: Craniofacial Frontal Analysis: Joseph G. Ghafari

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Posteroanterior Cephalometry: Craniofacial Frontal Analysis: Joseph G. Ghafari

Uploaded by

Copyright:

Available Formats

chapter 23

Complete Cephalometric poor radiographic technique, concern about additional

Frontal Analysis Structures and landmarks

Fig 23-2 Posteroanterior cephalometric tracing with landmarks

Correspondence of images of basic cranial and facial Evaluation of symmetry

15 cm, magnification with digital PA radiographs is about

Fig 23-9 Head tilt affects vertical measure-

Fig 23-10 Cephalic and facial indices. The

Leptoprosopic Mesoprosopic Euryprosopic

rate of growth of maxillary width apparently slows down

Normative data The distance CO-CO, a surrogate measure of cranial

Table 23-10 Comparison of selected angular measure-

Fig 23-13 Basic areas of differentiation in craniofacial development

Fig 23-14 Relation between maxillary width (J-J)

Fig 23-15 Relation between maxillary width (J-J)

expected increase in mandibular width with growth. In this

tions at both ages. This finding might conceptually support

1. The image to head size is displayed in a ratio of 1:1.

You might also like