See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/44617859

Craniofacial Growth, Maturation, and Change

Article  in  The Journal of craniofacial surgery · March 2010

DOI: 10.1097/SCS.0b013e3181cfea34 · Source: PubMed

CITATIONS READS

16 73

2 authors:

Ann H Ross Shanna Elizabeth Williams

North Carolina State University University of South Carolina School of Medicine - Greenville
92 PUBLICATIONS   1,050 CITATIONS    14 PUBLICATIONS   93 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Juvenile Taphonomy View project

Carnivore Scavenging in the Southeast: Implications for Estimating Time-Since-Death View project

All content following this page was uploaded by Ann H Ross on 25 July 2019.

The user has requested enhancement of the downloaded file.

 SCIENTIFIC FOUNDATION
Craniofacial Growth, Maturation, and Change:
Teens to Midadulthood
Ann H. Ross, PhD,* and Shanna E. Williams, PhDÞ
Abstract: Despite the attainment of several adult cranial dimen-
sions relatively early in childhood, skeletal maturity and, by con-
sequence, adult form are typically defined by the eruption of the
third molars around 17 years of age. This in turn serves as the
division between subadults and adults, which is then applied to
population studies of biological variation. Specifically, comparative
data sets of adult measurements are not directly applied to indi-
viduals who do not have complete skeletal growth, as it is believed
that the confounding effects of allometry may skew the results. The
present study uses geometric morphometrics techniques to in-
vestigate the appropriateness of this division with respect to three-
dimensional anatomical landmarks. Twenty-six landmarks were col-
lected from a single population of 24 crania partitioned into 4 age
groups spanning late adolescence to midadulthood. Generalized
Procrustes and multivariate statistical analyses were performed on
the landmark data. Results showed no significant morphological dif-
ferences between the teen and young adult age groups, whereas
significant shape and size differences were found in older adults re-
lative to their younger cohorts. Moreover, no growth-related shape
variation (ie, allometry) was detected within the sample. These find-
ings suggest that adult form is attained several years earlier than
commonly thought and corroborate other research that suggest
that subtle changes in cranial morphology continue throughout
adulthood.
Key Words: Subadult, morphological variation, geometric
morphometrics, cranial growth
(J Craniofac Surg 2010;21: 458Y461)
T he human cranium is composed of multiple structural units that
grow at different rates and in varying directions during postnatal
development. For instance, the neurocranium, which is primarily
influenced by brain growth, develops earlier and faster than the
facial skeleton during the early years of infancy and childhood.1Y3
From the *Department of Sociology and Anthropology, North Carolina State
University, Raleigh, North Carolina; and †Department of Anatomy and Cell
Biology, University of Florida, Gainesville, Florida.
Received August 12, 2009.
Accepted for publication November 7, 2009.
Address correspondence and reprint requests to Ann H. Ross, PhD,
Department of Sociology and Anthropology, North Carolina State
University, Raleigh, NC 27695-8107; E-mail: ann_ross@ncsu.edu
This project was funded by a National Institute of Justice grant
(No. 2005-MU-BX-K078).
Copyright * 2010 by Mutaz B. Habal, MD
ISSN: 1049-2275
DOI: 10.1097/SCS.0b013e3181cfea34
458 The Journal of Craniofacial Surgery
Copyright @ 2010 Mutaz B. Habal, MD. Unauthorized reproduction of this article is prohibited.
The slower-growing facial skeleton, meanwhile, is regulated by
the sensory organs and tooth eruption, which drive alveolar pro-
cess development.1Y3 Thus, fluctuations in tooth number or tooth
loss during the growth process directly affect the shape of the fa-
cial skeleton.2 Because of this differential growth, newborns and
infants exhibit a low and wide facial appearance. However, this
growth pattern shifts after age 4, with the neurocranium display-
ing the least relative growth of the structural units, the facial com-
ponents having an intermediate relative growth pattern, and the
mandibular components having the fastest relative growth.3 Al-
though craniofacial growth is considered complete around 17 years
after eruption of the permanent dentition, many regions of the cra-
nium achieve adult dimensions during early childhood. The length
of the cranium, which is 60% to 65% complete at birth, reaches
approximately 90% of its total size by 5 to 7 years of age.1 Cra-
nial width, meanwhile, is nearly 85% complete by 2 to 3 years,
and the cranium, as a whole, approaches overall adult size and
configuration by age 10 years.1,2
Despite the widespread attainment of adult craniofacial di-
mensions by the end of the first decade, researchers use eruption of
the third molars around age 17 as the marker by which adulthood is
distinguished from childhood, thus resulting in the relatively strict
partitioning of craniofacial data into adult (approximately older than
17 years) and subadult (approximately younger than 17 years) groups
during population studies. Further entrenched in this modus operandi
is the notion that biological constructs such as sex and ancestry
cannot be perceived in subadult skeletal material because they are
indistinct until stimulated by the differential growth of adolescence.
However, Strand Vi8arsdóttir and colleagues4 found that population-
specific facial morphologies are present at birth regardless of sex and
are modified during ontogeny, suggesting that ancestry can be deter-
mined in subadult cranial material. These findings indicate that the
characteristics of adulthood in the cranium may be acquired far
earlier than commonly believed. If true, this would allow for the
inclusion of subadults or some subset of this group into adult
population studies of craniofacial variation, which would not only
broaden our current understanding of human diversity but also
dramatically assist in the identification of subadult material because
comparative data sets of adult measurements could be applied to
individuals who have not completed skeletal growth. To further ex-
plore this notion, we present a preliminary study that uses geometric
morphometric techniques to compare cranial shape and size dif-
ferences from a single population of individuals spanning the early
teenage years to middle age.
MATERIALS AND METHODS
The total sample consists of 24 crania from the Morton Collec-
tion housed at the University of Pennsylvania Museum of Archaeology
and Anthropology. The individuals represent a native African popu-
lation enslaved in Cuba and were partitioned into 4 age categories:
14 years old (n = 4), 16 years old (n = 5), 20 years old (n = 3), and
25 years and older (age range, 25Y40 y; n = 12). Limited information is
known regarding age assignations, but they are thought to have been
& Volume 21, Number 2, March 2010
The Journal of Craniofacial Surgery & Volume 21, Number 2, March 2010 Craniofacial Growth: Teens to Adulthood

assigned by Dr Samuel Morton from known demographic information

in most cases.
Twenty-six three-dimensional type 1 and 2 anatomical land-
marks were collected from the crania using a hand-held digitizer
(Figs. 1 and 2; Table 1). Landmark selection was based on ease of
identification and collection and attempted to capture the general shape
of the craniofacial skeleton. Only type 1 and 2 landmarks were used, as
type 3 landmarks have been associated with a significant amount of
interobserver and intraobserver errors.5
Size and shape variations were analyzed separately within
the sample. Shape variation was accessed by subjecting the raw
landmark data to generalized Procrustes analysis (GPA) using the
morphometrics package of Morpheus et al (State University of
New York, Stony Brook). The GPA is an iterative least squares
procedure that optimally scales, rotates, and translates the data into a
common coordinate system. A principal component analysis (PCA)
FIGURE 2. Anatomic landmarks in the lateral view. The basion
of the covariance matrix was conducted on the GPA-transformed
and the opisthion, which represent the midpoint of the
shape variables to reduce the dimensionality of the data so as to meet
anterior and posterior margins of the foramen magnum,
the requirements of a parametric test. A multivariate analysis of
respectively, are not shown.
variance (MANOVA) was then performed using the PC scores to
evaluate shape differences among the age groups. Contrast tests were
also carried out within the MANOVA to detect the specific age among the age categories (F30,32.96 = 1.92; Pr 9 0.0355). However,
groups that differed in shape. when contrasted against one another, only the 20 and 25 years and
Size was computed as centroid size (CS), which is a measure older age groups displayed significant shape variation (F10,11 = 3.64;
of geometric scale. It is calculated as the square root of the sum of Pr 9 0.0224). Significant size differences, meanwhile, were ob-
squared distances of a set of landmarks from their centroid (mean). served between all of the younger age categories and the 25 years
In the absence of allometry, it is the only size measure not correlated and older age group (Table 2). Interestingly, neither the 14- nor the
with the shape variables.6 An independent-group t test was per- 16-year-old mean CSs were found to be significantly different from
formed to compare the mean CS of each age category. the 20-year-old CS. Finally, the Pearson product moment correlation
Although GPA scales all specimens to the same unit CS, it coefficients did not detect a significant correlation between the PC
does not correct for size-correlated shape changes (ie, allometry). To scores and the CS for the sample as a whole, meaning no significant
establish the independence of shape and size variations, allometric allometric effects were present within the sample (Table 3).
effects within the sample were examined by exploring the asso- Age-specific shape differences were visualized using vector
ciation of PC scores with CS. These associations were measured plots, which depict difference vectors extending from the landmarks of
using the Pearson product moment correlation coefficient (r). All one age group and going to the position of the same landmarks of
statistical analyses were conducted using SAS for Windows version another age group. These vectors represent the direction and mag-
9.1.3 (SAS, Cary, NC). nitude of difference between one mean age group form and another.
Figure 3 displays the consensus configuration for the 20-year-old age
RESULTS
Using the first 10 PC scores, which correspond to 83% of the
total variance, the MANOVA detected significant shape differences
TABLE 1. List of Type 1 and 2 Craniofacial Landmarks
Landmarks Abbreviations
Asterion left/right ast l/r
Basion bas
Bregma brg
Dacryon left/right dac l/r
Ectoconchion left/right ect l/r
Frontomalare temporale left/right fmt l/r
Glabella glb
Lambda lam
Mastoidale left/right mast l/r
Superior orbital height left/right obhs
Inferior orbital height left/right obhi
Opisthion ops
Prosthion proH
Subspinale ssp
Zygoorbitale left/right zygoo l/r
Zygomaxillare left/right zygom l/r
FIGURE 1. Anatomic landmarks in the anterior view.

* 2010 Mutaz B. Habal, MD 459

Copyright @ 2010 Mutaz B. Habal, MD. Unauthorized reproduction of this article is prohibited.
Ross and Williams The Journal of Craniofacial Surgery & Volume 21, Number 2, March 2010

TABLE 2. t Tests Comparing the Mean CSs

Age Group, y df t P
14 vs 16 7 0.11 0.919
14 vs 20 5 j1.32 0.246
14 vs 25 14 j4.42 0.0006
16 vs 20 6 j1.47 0.191
16 vs 25 15 j4.92 0.0002
20 vs 25 13 j2.86 0.013
Bold values show significance.

category deformed into the 25 years and older age category in the
anterior view. The dissimilarity between the 2 is expressed by the
length of the vectors, which have been magnified by a factor of 4.
In the 20-year-old age category, the landmark glabella is more pos-
teriorly oriented; frontomalare temporale, superior orbital border, and
ectoconchion are more medially oriented; dacryon is more supero-
medially placed; and prosthion is more inferoposteriorly placed. These
alterations suggest that an expansion in facial breadth and width oc-
curs with advancing age in adults.
In the lateral view (Fig. 4), the bregma has a more posterior
orientation, the lambda is more superior, the asterion is more
anteriorly placed, the mastoidale has a more inferior placement, and
basion is more anteriorly oriented in the 20-year-old age category
relative to the 25 years and older age category. These changes
indicate cranial base expansion with the adult aging process.
DISCUSSION AND CONCLUSIONS
The observed patterns of shape and size variations documen-
ted in this study, although preliminary, suggest that the chronological
age at which adult cranial form is attained and its stability thereafter is
far less straightforward than previously thought. The cessation of
craniofacial skeletal growth and initiation into adulthood is epito-
mized by the eruption of the permanent dentition around 17 years of
age. As such, measurements taken from individuals who have not yet
reached this dental stage traditionally are not combined with those
who have, so as to avoid the presumed confounding influence of
ontogenetic allometric changes occurring during the growth period.
This study, however, found no significant craniofacial shape or size
differences between 14-, 16-, and 20-year-olds, suggesting that
subadults reach their final form earlier than expected, and established
FIGURE 3. Anterior view of difference vectors showing the
dissimilarity between the consensus configurations of the
20 and 25 years and older age categories
(magnified by a factor of 4).
criteria regarding the minimal age at which adult cranial form is
attained is unnecessarily prohibitive.
In addition, despite Enlow’s1 suggestion that bone alteration
effectively ceases after adult cranial form is achieved, significant
size and shape differences were consistently detected involving the
25 years and older age group relative to the other age groups. No
allometric effects were present within the sample, and no significant
size variation was found between the 14-, 16-, and 20-year-old age
groups. However, the cranial size of the 25 years and older age group
was significantly different from those of all of the other younger age
groups. Moreover, the 20 and 25 years and older age groups displayed
significant shape differences, which concur with a study by Sarnäs and
Solow7 who found displacements in nasion and sella between 21 and
26 years of age. In their cephalometric study of craniofacial changes
in the third decade of life, Akgül and Toygar8 also found significant

TABLE 3. Pearson Product Moment Correlation Coefficients

of PCs and CSs
PC CS, r/P
1 j0.37/0.07
2 j0.17/0.41
3 0.15/0.47
4 0.15/0.47
5 j0.10/0.64
6 0.16/0.45
7 0.04/0.85
8 0.09/0.67 FIGURE 4. Lateral view of the difference vectors showing
9 0.23/0.30 the dissimilarity between the consensus configurations
10 0.02/0.48 of the 20 and 25 years and older age categories
(magnified by a factor of 4).

460 * 2010 Mutaz B. Habal, MD

Copyright @ 2010 Mutaz B. Habal, MD. Unauthorized reproduction of this article is prohibited.
The Journal of Craniofacial Surgery & Volume 21, Number 2, March 2010
changes, which were more pronounced in women and in the
lower region of the face. More recently, using semilandmark data,
Williams9 noted age-related differences in the curvature of the orbits,
zygomatic arches, and maxillary ridge of adult African and European
Americans.
The vector plots shown earlier (Figs. 3 and 4) reveal changes
in the face, vault, and base indicative of an overall pattern of cranial
expansion well into adulthood. Although Enlow1 is a major pro-
ponent of structural stasis after skeletal maturity, he does suggest
that remodeling can be stimulated in the presence of an intrinsic
environmental alteration such as a change in function, biomechan-
ical condition, and/or loss of dentition. Thus, these craniofacial
changes are in all likelihood multifactorial in nature and may be
related in part to the eruption of the third molars, to increasing size
of paranasal sinuses, and/or possibly to alveolar remodeling with
the aging process. In this study, the spatial modification observed in
basion, for instance, could be indicative of the prolonged growth of
the spheno-occipital synchondrosis into early adulthood. Function-
ally, this would allow for the posterior expansion of the maxilla
to accommodate the third molars and create space for the growing
nasopharynx.
In conclusion, this study uses geometric morphometric tech-
niques to explore morphological stability in late adolescence. Al-
though additional work is needed to further tease apart this issue, our
results suggest that adult form is reached during adolescence (ie,
early teens). Moreover, structural stasis is not inherent after skeletal
maturity, as marked morphological differences exist between early
adulthood and middle age. Incorporating such information into anal-
yses of population variation may not only assist in the identification
of unknown subadult skeletal material by allowing for the use of
* 2010 Mutaz B. Habal, MD
View publication stats Copyright @ 2010 Mutaz B. Habal, MD. Unauthorized reproduction of this article is prohibited.
Craniofacial Growth: Teens to Adulthood
adult comparative data sets but also create a more integrated model
of human diversity across age ranges than is currently possible.
ACKNOWLEDGMENT
The authors thank Dr Janet Monge for assistance with the
Morton Collection.
REFERENCES
1. Enlow DH. Facial Growth. 3rd ed. Philadelphia: WB Saunders
Company, 1990
2. Schumacher G-H. Principles of skeletal growth. In: Dixon A, Hoyte D,
Rönning O, eds. Fundamentals of Craniofacial Growth. Boca Raton,
FL: CRC Press, 1997:21
3. Sperber GH. Craniofacial Development. London: BC Decker Inc, 2001
4. Vi8arsdóttir UN, O_Higgins P, Stringer C, et al. A geometric
morphometric study of regional differences in the ontogeny of the
modern human facial skeleton. J Anat 2002;201:211Y229
5. Ross AH, Williams S. Testing repeatability and error of coordinate
cranial landmark data acquired from skulls. J Forensic Sci
2008;53:782Y785
6. Bookstein FL. Morphometrics Tools for Landmarks Data. Geometry
and Biology. New York: Cambridge University Press, 1991
7. Sarnäs KV, Solow B. Early adult changes in the skeletal and
soft-tissue profile. Eur J Orthod 1980;2:1Y12
8. Akgül AA, Toygar TU. Natural craniofacial changes in the third
decade of life: a longitudinal study. Am J Orthod Dentofacial Orthop
2002;122:512Y522
9. Williams SE. Is Aging Only Skin Deep? Assessing Change in Facial
Bone Curvature With Age [dissertation]. Gainesville: University of
Florida, 2008
461