PAPER
J Forensic Sci, March 2015, Vol. 60, No. 2 doi: 10.1111/1556-4029.12645 Available online at: onlinelibrary.wiley.com
ANTHROPOLOGY John M. Starbuck,1 Ph.D.; Ahmed Ghoneima,1 Ph.D.; and Katherine Kula,1 DMD
Facial Tissue Depths in Children with Cleft Lip and Palate*
ABSTRACT: Cleft lip and palate (CLP) is a craniofacial malformation affecting more than seven million people worldwide that results in defects of the hard palate, teeth, maxilla, nasal spine and floor, and maxillodental asymmetry. CLP facial soft-tissue depth (FSTD) values have never been published. The purpose of this research is to report CLP FSTD values and compare them to previously published FSTD values for normal children. Thirty-eight FSTDs were measured on cone beam computed tomography images of CLP children (n = 86; 7–17 years). MANOVA and ANOVA tests determined whether cleft type, age, sex, and bone graft surgical status affect tissue depths. Both cleft type (unilateral/bilateral) and age influence FSTDs. CLP FSTDs exhibit patterns of variation that differ from normal children, particularly around the oronasal regions of the face. These differences should be taken into account when facial reconstructions of children with CLP are created.
KEYWORDS: forensic science, forensic anthropology, facial soft-tissue depth, forensic facial reconstruction, facial reproduction, facial restoration, cleft lip and cleft palate, cone beam computed tomography
Facial reconstruction is the process of rebuilding the facial morphology of a deceased individual upon the skull (1,2). Facial reconstruction is carried out manually using clay-based reconstructions or digitally with computer programs (3–8). In medicolegal investigations, facial reconstructions are typically circulated to the public as a last resort to stimulate recognition of a deceased individual when other forensics methods (e.g., DNA analysis, dental record assessment) have proven fruitless (9–11). Facial reconstruction may be a useful tool for identification in situations where large numbers of individuals have died due to terrorism, acts of war, and natural disasters (1). Moreover, facial reconstruction can be useful in archaeological and paleoanthropological contexts to estimate the facial morphology of historical figures and purported human ancestors (12–18). Age-related changes in facial proportions, tissue thickness, and facial appearance occur as individuals grow and throughout senescence (19–28). Several studies of facial soft-tissue depth (FSTD) have focused on children with different ethnic backgrounds and skeletal classes including: Caucasian (29–32), African–American (29,33), Japanese (34,35), Hispanic (29), Class I, II, and III occlusion patterns (30,34,36), and a meta-analysis of pooled FSTD data (37). Because of the high measurement error (i.e., statistical noise) and skewed data associated with previous studies of tissue depth, it has been argued that subadult FSTDs should only be divided into younger and older age groups (37). 1 Department of Orthodontics and Oral Facial Genetics, School of Dentistry, Indiana University, Indianapolis, IN 46202. *Funding provided by the Indiana University Signature Center Initiative: 3D Imaging of the Craniofacial Complex Center; Jarabak Endowed Professorship Received 27 June 2013; and in revised form 2 Feb. 2014; accepted 6 Mar. 2014.
274
Cleft lip and palate (CLP) is a relatively common craniofacial birth defect affecting 1 in 1000 children per year (38–40), which equates to more than seven million contemporary people worldwide. Today, the lips and alveoli of children with clefts are surgically repaired early in life (41). The combination of congenital clefts and subsequent functional and esthetic surgical grafts/corrections can result in changes to nasal septum growth, malocclusion, maxillodental asymmetry, reductions of nasal airway size, defects in upper lip musculature, and increased reliance on mouth breathing (42–50). These factors may affect FSTDs by causing scarring, muscle pull, changes to nasal form and function, and tissue thickness changes around the site of the congenital cleft and subsequent surgical repair (51–55). Even after surgical repair, a suite of diagnostic differences are present on the craniofacial skeleton (e.g., deviated nasal septum, missing lateral incisors, misaligned teeth, missing or perforated nasal floor, bone scalloping on the alveolus, reduced nasal airway volume, and asymmetry of the anterior nasal spine, nasal floor, nasal opening, and maxilla) that allow forensic investigators to determine that the individual was born with CLP (52,54,55) (Fig. 1). Tissue depths for children with surgically corrected CLP have not been previously published. The primary goal of this study is to report FSTD values that were measured from 3D cone beam computed tomography (CBCT) images of children with surgically corrected CLP and to evaluate these values statistically to determine whether cleft type, age, sex, or the use of bone grafts during surgical correction have significant effects on tissue depth values. The secondary goal is to compare CLP FSTD values to previously published pooled tissue depths for normal children (37,56) to determine how patterns of FSTD variation differ across the faces of individuals born with CLP. © 2014 American Academy of Forensic Sciences
STARBUCK ET AL.
.
FACIAL TISSUE DEPTHS IN CHILDREN
275
FIG. 1––Midfacial morphology of the craniofacial skeleton of individuals who have surgically corrected cleft lip and palate (CLP). Even after surgical repair, which may or may not include an alveolar bone graft, individuals born with CLP continue to exhibit a diagnostic suite of morphological characteristics that can include a deviated nasal septum, missing lateral incisors, missing or perforated nasal floor, bone scalloping on the alveolus, reduced nasal airway volume, and asymmetry of the anterior nasal spine, nasal floor, nasal opening (outlined), and maxilla. (A) An individual born with unilateral CLP who was not surgically treated with an alveolar bone graft. Despite other types of surgical correction, nasal asymmetry and a partial cleft are readily visible. (B) An individual born with bilateral CLP who was treated with an alveolar bone graft. Heavy bone scalloping and misalignment of maxillary teeth are clearly visible. (C) An individual born with unilateral CLP who was treated with an alveolar bone graft. The anterior nasal spine, nasal floor, and nasal opening are asymmetric.
TABLE 1––Samples used in cleft lip and palate (CLP) facial soft-tissue depth (FSTD) analyses.
Sample Sample Sample Sample
1 2 3 4
Diagnosis
Sample Size (N = 86)
Number of Males (n = 61)
Number of Females (n = 25)
Unilateral CLP Bilateral CLP Unilateral CLP Bilateral CLP
35 19 20 12
25 13 15 8
10 6 5 4
Materials and Methods Cone beam computed tomography images (0.3–0.4 mm voxel size; i-Cat machine, Imaging Sciences International LLC, Hatfield, PA) were obtained for this study from pre-existing orthodontic records at the Indiana University School of Dentistry with IRB approval from the Indiana University Human Subjects Office (Study # 1210009813). To limit unnecessary radiation exposure only one CBCT image was acquired of each individual. Inclusion criteria required children to be previously diagnosed with Veau Class III unilateral CLP or Veau Class IV bilateral CLP that was surgically repaired. Veau Class III and IV patients exhibit clefting of both the soft and hard tissues of the lip, alveolus, and palate. Based on these criteria, 86 children were included in this study (Table 1). Due to limitations of sample size, no attempt was made to control for ancestry; however, the majority of the sample was Caucasian. Approximately 48% of these individuals previously had an alveolar bone graft as part of their surgical treatment algorithm. For graphical and statistical comparisons, the overall sample was divided into subsamples
Mean Age (years) Standard Deviation 9 9 14 15
1.33 1.06 1.46 1.85
Age Range (years) 7–11 7–11 12–17 12–17
based on CLP type (unilateral or bilateral), age, age group (7–11 years or 12–17 years), sex, and bone graft status. Coded CBCT images were analyzed using 3D Dolphin Imaging software (v11.5; Chatsworth, CA) by the same individual (JMS) on the same computer and monitor. Orientation was standardized using a 3D orientation module in Dolphin software that positioned CBCT images in lateral view by passing a line through orbitale and porion, and in frontal view by passing a line through nasion and pogonion. Thirty-eight FSTD values were collected from each CBCT image by locating midline and bilateral hard- and soft-tissue landmarks on the skull and face (Table 2, Fig. 2) using Dolphin software and integrated measurement tools. All statistical testing was conducted using Minitab, State College, PA, USA (v.16.1.0) or SPSS, Armonk, NY, USA (v.20.0.0.1). During a repeatability study, all FSTDs (Table 2) were measured on 10 randomized coded images on three separate occasions with at least 24 hours between each measurement session to avoid memory bias. The order or individuals measured was randomized in Excel using a randomization function (i.e.,
Midline points 1. Supraglabella (sg)—The most anterior point of the forehead, superior to glabella in the midsagittal plane. 2. Glabella (g)—The most prominent point between the supra orbital ridges in the midsagittal plane. 3. Nasion (n)—Superior midline intersection of the nasal bones. 4. Rhinion (rhi)—The anterior tip of the nasal bones. 5. Mid-philtrum (mp)—Midline of the maxilla, midway between the base of the nasal spine and the anterior edge of the maxilla along the sagittal plane. 6. Labrale superius (ls)—Midline at the most anterior edge of the superior alveolar ridge of the maxillae, centered between the maxillary central incisors at the level of the cementum–enamel junction. 7. Labrale inferius (li)—Midline at the most inferior edge of the inferior alveolar ridge of the mandible, centered between the mandibular central incisors at the level of the cementum–enamel junction. 8. Mentolabial sulcus (mls)—Deepest midline point in the groove superior to the eminence. 9. Pogonion (pg)—Most anterior midline point on the eminence of the mandible. 10. Menton (m)—Most inferior midline point at the symphysis of the mandible. Bilateral points 11. Frontal eminence (fe)—Bony projection of the ectocranial surface of the frontal bone. 12. Mid-Supraorbital (mso)—Centered superior margin of the orbit. 13. Mid-Infraorbital (mio)—Centered inferior margin of the orbit. 14. Lateral orbit (lo)—Center of the zygomatic process lined up with the lateral border of the orbit along the anterior—posterior axis. 15. Nasal Ala Furrow (naf)—Lateral border of the nasal aperture located directly underneath soft-tissue nasal ala furrow (naf’). 16. Subalare (sbal)—Inferior border of the nasal aperture located directly underneath soft-tissue subalare (sbal’). 17. Mediolateral philtrum (mlp)—Point located where a sagittal line passes through the midpoint between the widest point of the nasal aperture and the anterior nasal spine and intersects with an axial line passing midway through the upper jaw above the teeth and below the inferior nasal aperture. 18. Lateral philtrum (lp)—Point located where a sagittal line passes through the widest point of the nasal aperture and intersects with an axial line passing midway through the upper jaw above the teeth and below the inferior nasal aperture. 19. Zygion (zy)—The most lateral point of the zygomatic arch. 20. Supraglenoid (sgl)—Root of the zygomatic arch just before the ear. 21. Gonion (go)—Point located on the jaw line at the level of the angle between the posterior and the inferior borders of the mandible. 22. Supra M1 (sm1)—Point located on the alveolar process at the level of the middle of the first upper molar. 23. Occlusal line (ol)—Point located on anterior margin of the ramus of the mandible, in alignment with the plane of dental occlusion. 24. Infra M1 (im1)—Point located on the alveolar process at the level of the middle of the first lower molar.
Skeletal Landmarks
Midline points 1. Supraglabella (sg’)—The midline soft-tissue point directly overlying hard-tissue supraglabella (sg). 2. Glabella (g’)—The midline soft-tissue point directly overlying hard-tissue glabella (g). 3. Nasion (n’)—The midline soft-tissue point directly overlying hard-tissue nasion (n). 4. Rhinion (rhi’)—The midline soft-tissue point directly overlying hard-tissue rhinion (rhi). 5. Mid-philtrum (mp’)—The midline soft-tissue point midway between subnasale and the vermilion border of the upper lip. 6. Labrale superius (ls’)—The midline soft-tissue point at labiale superius on the vermilion border of the upper lip. 7. Labrale inferius (li’)—The midline soft-tissue point at labiale inferius on the vermilion border of the lower lip. 8. Mentolabial Sulcus (mls’)—The midline soft-tissue point directly overlying hard-tissue supramentale (sm). 9. Pogonion (pg’)—The midline soft-tissue point on the eminence of the soft-tissue chin overlying hard-tissue pogonion (pg). 10. Menton (m’)—The midline soft-tissue point directly overlying hard-tissue menton (m). Bilateral points 11. Frontal eminence (fe’)—The soft-tissue point directly overlying hard-tissue frontal eminence (fe). 12. Mid-Supraorbital (mso’)—The soft-tissue point directly overlying hard-tissue supraorbital (so). 13. Mid-Infraorbital (mio’)—The soft-tissue point directly overlying hard-tissue infraorbital (io). 14. Lateral orbit (lo’)—The soft-tissue point directly overlying hard-tissue lateral orbit (lo). 15. Nasal Ala Furrow (naf’)—Superior insertion of soft-tissue nasal ala furrow into lateral nose. 16. Subalare (sbal’)—The point at the lower limit of each alar base, where the alar base disappears into the skin of the upper lip. 17. Mediolateral philtrum (mlp’)—The soft-tissue point directly overlying hard-tissue mediolateral philtrum (mlp). 18. Lateral philtrum (lp)—The soft-tissue point directly overlying hard-tissue lateral philtrum (lp). 19. Zygion (zy’)—The soft-tissue point directly overlying hard-tissue zygomatic arch (za). 20. Supraglenoid (sgl’)—The soft-tissue point directly overlying hard-tissue supraglenoid (sg). 21. Gonion (go’)—The soft-tissue point directly overlying hard-tissue gonion (go). 22. Supra M1 (sm1’)—The soft-tissue point directly overlying hard-tissue supra m1 (sm1). 23. Occlusal line (ol’)—The soft-tissue point directly overlying hard-tissue occlusal line (ol). 24. Infra M1 (im1’)—The soft-tissue point directly overlying hard-tissue infra m1 (im1).
Corresponding Soft-Tissue Landmarks
TABLE 2––Skeletal- and soft-tissue anatomical landmarks used to measure facial soft-tissue depths (FSTDs) from cone beam computed tomography (CBCT) images.
276 JOURNAL OF FORENSIC SCIENCES
STARBUCK ET AL.
.
FACIAL TISSUE DEPTHS IN CHILDREN
277
FIG. 2––Anatomical landmarks located on cone beam computed tomography (CBCT) images to measure facial soft-tissue depths (FSTDs). Dolphin software was used to automatically calculate tissue depths between the hard and soft-tissue anatomical landmarks (shown overlaid upon each other here). Hard and soft-tissue landmark definitions are provided in table 2.
TABLE 3––Technical error of measurement (TEM) across each permutation of the three measurement trials. Midline FSTDs (mm) TEM trial 1 versus trial 2 TEM trial 2 versus trial 3 TEM trial 1 versus trial 3 Mean TEM Lateral FSTDs (mm) TEM trial 1 versus trial 2 TEM trial 2 versus trial 3 TEM trial 1 versus trial 3 Mean TEM
sg–sg’
g–g’
n–n’
rhi–rhi’
mp–mp’
ls–ls’
li–li’
mls–mls’
pg–pg’
m–m’
0.13
0.15
0.10
0.13
0.19
0.24
0.32
0.23
0.21
0.23
0.13
0.09
0.14
0.11
0.21
0.30
0.11
0.21
0.19
0.18
0.19
0.10
0.11
0.08
0.15
0.16
0.30
0.27
0.09
0.11
0.15 fe–fe’
0.12 mio– mio’ 0.22
0.11 lo–lo’
0.19 naf–naf’
0.23 sbal–sbal’
0.25 mlp–mlp’
0.24 lp–lp’
0.16 zy–zy’
0.17 sgl–sgl’
go–go’
0.22
0.11 mso– mso’ 0.17
0.25
0.17
0.21
0.12
0.13
0.14
0.22
0.20
0.13
0.15
0.31
0.19
0.16
0.12
0.14
0.12
0.18
0.17
0.20
0.19
0.14
0.19
0.11
0.12
0.20
0.16
0.19
0.25
0.17
0.18
0.12
0.13
ol–ol’
0.19
sm1– sm1’ 0.21
0.24
im1– im1’ 0.37
0.28
0.19
0.29
0.17
0.24
0.10
0.18
0.17
0.29
0.32
0.35
0.12
0.23
0.18
0.27
0.24
0.32
Overall Mean TEM = 0.19 mm.
TABLE 4––Midline mean facial soft-tissue depth (FSTD) values (mm) for unilateral and bilateral cleft lip and palate (CLP) individuals. Midline Tissue Depths Tissue Depth Measure (mm)
sg–sg’
Unilateral CL/CP (n = 35), Ages 7–11 Average 3.9 Standard Deviation 0.8 Median 3.9 Bilateral CL/CP (n = 19), Ages 7–11 Average 4.4 Standard Deviation 1.2 Median 4.0 Unilateral CL/CP (n = 20), Ages 12–17 Average 5.4 Standard Deviation 1.1 Median 5.6 Bilateral CL/CP (n = 12), Ages 12–17 Average 4.9 Standard Deviation 0.9 Median 4.8
g–g’
n–n’
rhi–rhi’
mp–mp’
ls–ls’
li–li’
mls–mls’
pg–pg’
m–m’
4.9 0.9 4.8
6.2 1.2 6.2
3.4 0.9 3.2
10.9 2.3 10.7
11.7 2.4 11.1
10.9 2.5 10.6
10.4 2.0 10.1
8.9 2.2 9.0
6.9 2.4 6.9
5.6 1.4 5.3
6.7 1.4 6.6
3.4 0.8 3.1
10.7 2.0 10.4
10.7 1.7 10.4
11.9 3.4 12.0
10.6 2.5 10.9
9.4 2.4 8.4
7.2 1.9 6.9
6.1 1.1 6.2
7.3 1.5 7.3
4.4 1.4 4.3
12.1 2.2 12.6
14.3 2.4 14.0
12.3 3.0 11.8
12.1 1.9 12.1
10.8 2.1 10.8
8.2 2.6 7.6
5.9 0.9 6.1
7.1 1.0 7.2
3.9 1.0 3.8
12.2 2.6 12.8
14.2 3.6 14.5
14.8 2.9 15.7
12.9 1.8 12.9
11.0 2.5 11.2
8.0 2.3 7.9
278
JOURNAL OF FORENSIC SCIENCES TABLE 5––Lateral mean facial soft-tissue depth (FSTD) values (mm) for unilateral and bilateral cleft lip and palate (CLP) individuals.
Bilateral Tissue Depths Tissue Depth Measure (mm)
fe–fe’
mso–mso’
mio–mio’
Unilateral CL/CP (n = 35), Ages 7–11 Average 4.4 6.1 6.7 Standard 1.1 1.4 1.7 Deviation Median 4.3 6.1 6.4 Bilateral CL/CP (n = 19), Ages 7–11 Average 4.9 6.9 7.5 Standard 1.0 2.0 2.0 Deviation Median 4.8 6.5 7.2 Unilateral CL/CP (n = 20), Ages 12–17 Average 5.5 8.3 7.5 Standard 1.2 2.8 1.7 Deviation Median 5.6 7.9 7.9 Bilateral CL/CP (n = 12), Ages 12–17 Average 5.9 8.1 8.0 Standard 1.5 1.5 2.6 Deviation Median 5.5 8.0 8.0
lo–lo’
naf–naf’
sbal–sbal’
mlp–mlp’
lp–lp’
zy–zy’
sgl–sgl’
go–go’
sm1–sm1’
ol–ol’
im1–im1’
12.4 2.7
13.5 2.5
12.4 2.3
12.1 2.7
12.6 2.8
8.1 1.9
10.4 2.4
11.4 3.2
28.4 3.8
20.2 2.7
22.7 3.3
12.4
13.2
12.4
12.0
12.4
7.8
10.0
11.0
29.1
19.8
22.9
13.9 2.8
14.5 2.1
14.8 2.6
11.1 2.6
14.9 3.1
9.5 2.7
10.1 1.6
13.2 5.1
29.4 3.9
20.7 3.7
23.7 4.4
13.6
14.6
14.5
11.1
14.9
8.8
9.9
12.2
27.9
20.0
22.3
13.9 3.2
18.3 7.8
14.8 2.9
13.6 2.7
14.9 2.7
9.4 2.5
11.8 3.1
13.7 4.7
30.1 5.1
21.7 3.8
24.5 4.9
14.4
17.0
15.0
13.8
14.4
8.7
11.5
13.0
31.0
21.6
24.7
13.4 4.1
16.8 2.8
15.5 2.6
13.5 3.1
15.9 2.2
9.4 3.0
11.0 2.2
14.7 6.6
33.6 7.7
24.4 6.7
27.4 7.0
14.0
17.7
15.6
12.3
16.5
9.4
11.0
12.1
33.6
22.0
25.3
ity across three repeatability trials. The ICC value was 0.99 and indicated that FSTD measurements were strongly reliable. The technical error of measurement (TEM) was calculated for each permutation of measurement error trials to assess intra-observer error using the following: rP ffiffiffiffiffiffiffiffiffiffiffiffi D2 ; TEM ¼ 2N
FIG. 3––Mean facial soft-tissue depth (FSTD) comparison (mm) among age groups born with the same cleft lip and palate (CLP) phenotype. (A) FSTDs among the two unilateral CLP age groups (dashed and solid line). (B) FSTDs among the two bilateral CLP age groups (dashed and solid line).
=RAND()*100) and sorting the randomly generated numbers and coded image column numbers. The intraclass correlation (ICC) coefficient was calculated to assess intra-observer reliabil-
where D was equal to the difference of measured values between two trials, and N was equal to the number of individuals measured (17). TEM values for each FSTD measurement comparison are provided (Table 3) with the overall mean TEM of 0.19 mm being considered adequately low for the purposes of this investigation. As previous investigations have noted (37,56), it is especially important to minimize measurement error from imaging artifacts and human error. Individuals in this study were only imaged once to minimize radiation exposure to living subjects, thereby precluding the assessment of measurement error associated with the CBCT i-Cat machine. As a result, it is possible that errors are larger than those reported. After assessing reliability and TEM, FSTDs (Fig. 2) were measured on each image on three separate occasions with at least 24 hours between each measurement session. FSTD measurements that differed by more than 1 mm from homologous FSTD values collected from the same individual during different trials were remeasured and replaced to minimize measurement error occurring from human error and imaging artifacts. Individual FSTD values for each age group were tested for normal distributions using a Kolmogorov–Smirnov test for normality (a ≤ 0.05). Scatterplots and regression lines were explored for FSTD values and age. Average FSTD values and previously published pooled FSTD values for children (37) were graphed to explore similarities and differences. Data were analyzed using a multivariate analysis of variance (MANOVA)
STARBUCK ET AL.
.
FACIAL TISSUE DEPTHS IN CHILDREN
279
FIG. 4––Scatterplots and regression lines for facial soft-tissue depths (FSTDs) (y-axis in mm) and age (x-axis in years) combined for unilateral and bilateral cleft lip and palate (CL/CP) samples.
followed by separate post hoc ANOVA tests for independent variables showing significant effects on the dependent variables. MANOVA is a statistical test for assessing the null hypothesis of the equality of multiple averages and is used to determine whether one or more independent variables have a significant effect on two or more dependent variables. MANOVA is appropriate here because FSTDs represent multivariate data measured from the facial complex, and this method is capable of detecting statistical differences that may not be discovered when only using univariate statistics. The independent variables used in the MANOVA analysis are cleft type (unilateral/bilateral), age (7– 17 years), age group (younger 7–11 years/older 12–17 years), sex (male/female), and bone graft status (yes/no), and the dependent variables are all of the unilateral and averaged bilateral FSTD values. Post hoc ANOVA tests were carried out to determine which FSTDs were significantly affected by independent
variables shown to be significant by the MANOVA. Statistical significance was initially set at p-value ≤ 0.05, but lowered to p-value ≤ 0.016 after a Bonferonni adjustment for 3 tests (one MANOVA and two ANOVAs). Results Summary statistics for midline and lateral tissue depth values are provided in Tables 4 and 5. Average FSTD values for the unilateral and bilateral CL/CP age groups (i.e., 7–11 years and 12–17 years) are equivalent, and patterns of variation for the two age groups across the face are nearly identical to each other (Fig. 3A and B). Average FSTD values for younger (7–11 years) and older (12–17 years) age groups follow similar patterns of variation, although FSTDs in the older age group tend to be thicker (Fig. 3A and B). Scatterplots of tissue depths versus age
280
JOURNAL OF FORENSIC SCIENCES
FIG. 5––Scatterplots and regression lines for facial soft-tissue depths (FSTDs) (y-axis in mm) and age (x-axis in years) combined for unilateral and bilateral cleft lip and palate (CL/CP) samples (continued).
with fitted regression lines were created (Figs 4 and 5) and show that all FSTDs have positive slopes and tend to increase with age on average. A MANOVA revealed a significant multivariate main effect for cleft type (p-value = 0.034) and age (p-value
J Forensic Sci, March 2015, Vol. 60, No. 2 doi: 10.1111/1556-4029.12645 Available online at: onlinelibrary.wiley.com
ANTHROPOLOGY John M. Starbuck,1 Ph.D.; Ahmed Ghoneima,1 Ph.D.; and Katherine Kula,1 DMD
Facial Tissue Depths in Children with Cleft Lip and Palate*
ABSTRACT: Cleft lip and palate (CLP) is a craniofacial malformation affecting more than seven million people worldwide that results in defects of the hard palate, teeth, maxilla, nasal spine and floor, and maxillodental asymmetry. CLP facial soft-tissue depth (FSTD) values have never been published. The purpose of this research is to report CLP FSTD values and compare them to previously published FSTD values for normal children. Thirty-eight FSTDs were measured on cone beam computed tomography images of CLP children (n = 86; 7–17 years). MANOVA and ANOVA tests determined whether cleft type, age, sex, and bone graft surgical status affect tissue depths. Both cleft type (unilateral/bilateral) and age influence FSTDs. CLP FSTDs exhibit patterns of variation that differ from normal children, particularly around the oronasal regions of the face. These differences should be taken into account when facial reconstructions of children with CLP are created.
KEYWORDS: forensic science, forensic anthropology, facial soft-tissue depth, forensic facial reconstruction, facial reproduction, facial restoration, cleft lip and cleft palate, cone beam computed tomography
Facial reconstruction is the process of rebuilding the facial morphology of a deceased individual upon the skull (1,2). Facial reconstruction is carried out manually using clay-based reconstructions or digitally with computer programs (3–8). In medicolegal investigations, facial reconstructions are typically circulated to the public as a last resort to stimulate recognition of a deceased individual when other forensics methods (e.g., DNA analysis, dental record assessment) have proven fruitless (9–11). Facial reconstruction may be a useful tool for identification in situations where large numbers of individuals have died due to terrorism, acts of war, and natural disasters (1). Moreover, facial reconstruction can be useful in archaeological and paleoanthropological contexts to estimate the facial morphology of historical figures and purported human ancestors (12–18). Age-related changes in facial proportions, tissue thickness, and facial appearance occur as individuals grow and throughout senescence (19–28). Several studies of facial soft-tissue depth (FSTD) have focused on children with different ethnic backgrounds and skeletal classes including: Caucasian (29–32), African–American (29,33), Japanese (34,35), Hispanic (29), Class I, II, and III occlusion patterns (30,34,36), and a meta-analysis of pooled FSTD data (37). Because of the high measurement error (i.e., statistical noise) and skewed data associated with previous studies of tissue depth, it has been argued that subadult FSTDs should only be divided into younger and older age groups (37). 1 Department of Orthodontics and Oral Facial Genetics, School of Dentistry, Indiana University, Indianapolis, IN 46202. *Funding provided by the Indiana University Signature Center Initiative: 3D Imaging of the Craniofacial Complex Center; Jarabak Endowed Professorship Received 27 June 2013; and in revised form 2 Feb. 2014; accepted 6 Mar. 2014.
274
Cleft lip and palate (CLP) is a relatively common craniofacial birth defect affecting 1 in 1000 children per year (38–40), which equates to more than seven million contemporary people worldwide. Today, the lips and alveoli of children with clefts are surgically repaired early in life (41). The combination of congenital clefts and subsequent functional and esthetic surgical grafts/corrections can result in changes to nasal septum growth, malocclusion, maxillodental asymmetry, reductions of nasal airway size, defects in upper lip musculature, and increased reliance on mouth breathing (42–50). These factors may affect FSTDs by causing scarring, muscle pull, changes to nasal form and function, and tissue thickness changes around the site of the congenital cleft and subsequent surgical repair (51–55). Even after surgical repair, a suite of diagnostic differences are present on the craniofacial skeleton (e.g., deviated nasal septum, missing lateral incisors, misaligned teeth, missing or perforated nasal floor, bone scalloping on the alveolus, reduced nasal airway volume, and asymmetry of the anterior nasal spine, nasal floor, nasal opening, and maxilla) that allow forensic investigators to determine that the individual was born with CLP (52,54,55) (Fig. 1). Tissue depths for children with surgically corrected CLP have not been previously published. The primary goal of this study is to report FSTD values that were measured from 3D cone beam computed tomography (CBCT) images of children with surgically corrected CLP and to evaluate these values statistically to determine whether cleft type, age, sex, or the use of bone grafts during surgical correction have significant effects on tissue depth values. The secondary goal is to compare CLP FSTD values to previously published pooled tissue depths for normal children (37,56) to determine how patterns of FSTD variation differ across the faces of individuals born with CLP. © 2014 American Academy of Forensic Sciences
STARBUCK ET AL.
.
FACIAL TISSUE DEPTHS IN CHILDREN
275
FIG. 1––Midfacial morphology of the craniofacial skeleton of individuals who have surgically corrected cleft lip and palate (CLP). Even after surgical repair, which may or may not include an alveolar bone graft, individuals born with CLP continue to exhibit a diagnostic suite of morphological characteristics that can include a deviated nasal septum, missing lateral incisors, missing or perforated nasal floor, bone scalloping on the alveolus, reduced nasal airway volume, and asymmetry of the anterior nasal spine, nasal floor, nasal opening (outlined), and maxilla. (A) An individual born with unilateral CLP who was not surgically treated with an alveolar bone graft. Despite other types of surgical correction, nasal asymmetry and a partial cleft are readily visible. (B) An individual born with bilateral CLP who was treated with an alveolar bone graft. Heavy bone scalloping and misalignment of maxillary teeth are clearly visible. (C) An individual born with unilateral CLP who was treated with an alveolar bone graft. The anterior nasal spine, nasal floor, and nasal opening are asymmetric.
TABLE 1––Samples used in cleft lip and palate (CLP) facial soft-tissue depth (FSTD) analyses.
Sample Sample Sample Sample
1 2 3 4
Diagnosis
Sample Size (N = 86)
Number of Males (n = 61)
Number of Females (n = 25)
Unilateral CLP Bilateral CLP Unilateral CLP Bilateral CLP
35 19 20 12
25 13 15 8
10 6 5 4
Materials and Methods Cone beam computed tomography images (0.3–0.4 mm voxel size; i-Cat machine, Imaging Sciences International LLC, Hatfield, PA) were obtained for this study from pre-existing orthodontic records at the Indiana University School of Dentistry with IRB approval from the Indiana University Human Subjects Office (Study # 1210009813). To limit unnecessary radiation exposure only one CBCT image was acquired of each individual. Inclusion criteria required children to be previously diagnosed with Veau Class III unilateral CLP or Veau Class IV bilateral CLP that was surgically repaired. Veau Class III and IV patients exhibit clefting of both the soft and hard tissues of the lip, alveolus, and palate. Based on these criteria, 86 children were included in this study (Table 1). Due to limitations of sample size, no attempt was made to control for ancestry; however, the majority of the sample was Caucasian. Approximately 48% of these individuals previously had an alveolar bone graft as part of their surgical treatment algorithm. For graphical and statistical comparisons, the overall sample was divided into subsamples
Mean Age (years) Standard Deviation 9 9 14 15
1.33 1.06 1.46 1.85
Age Range (years) 7–11 7–11 12–17 12–17
based on CLP type (unilateral or bilateral), age, age group (7–11 years or 12–17 years), sex, and bone graft status. Coded CBCT images were analyzed using 3D Dolphin Imaging software (v11.5; Chatsworth, CA) by the same individual (JMS) on the same computer and monitor. Orientation was standardized using a 3D orientation module in Dolphin software that positioned CBCT images in lateral view by passing a line through orbitale and porion, and in frontal view by passing a line through nasion and pogonion. Thirty-eight FSTD values were collected from each CBCT image by locating midline and bilateral hard- and soft-tissue landmarks on the skull and face (Table 2, Fig. 2) using Dolphin software and integrated measurement tools. All statistical testing was conducted using Minitab, State College, PA, USA (v.16.1.0) or SPSS, Armonk, NY, USA (v.20.0.0.1). During a repeatability study, all FSTDs (Table 2) were measured on 10 randomized coded images on three separate occasions with at least 24 hours between each measurement session to avoid memory bias. The order or individuals measured was randomized in Excel using a randomization function (i.e.,
Midline points 1. Supraglabella (sg)—The most anterior point of the forehead, superior to glabella in the midsagittal plane. 2. Glabella (g)—The most prominent point between the supra orbital ridges in the midsagittal plane. 3. Nasion (n)—Superior midline intersection of the nasal bones. 4. Rhinion (rhi)—The anterior tip of the nasal bones. 5. Mid-philtrum (mp)—Midline of the maxilla, midway between the base of the nasal spine and the anterior edge of the maxilla along the sagittal plane. 6. Labrale superius (ls)—Midline at the most anterior edge of the superior alveolar ridge of the maxillae, centered between the maxillary central incisors at the level of the cementum–enamel junction. 7. Labrale inferius (li)—Midline at the most inferior edge of the inferior alveolar ridge of the mandible, centered between the mandibular central incisors at the level of the cementum–enamel junction. 8. Mentolabial sulcus (mls)—Deepest midline point in the groove superior to the eminence. 9. Pogonion (pg)—Most anterior midline point on the eminence of the mandible. 10. Menton (m)—Most inferior midline point at the symphysis of the mandible. Bilateral points 11. Frontal eminence (fe)—Bony projection of the ectocranial surface of the frontal bone. 12. Mid-Supraorbital (mso)—Centered superior margin of the orbit. 13. Mid-Infraorbital (mio)—Centered inferior margin of the orbit. 14. Lateral orbit (lo)—Center of the zygomatic process lined up with the lateral border of the orbit along the anterior—posterior axis. 15. Nasal Ala Furrow (naf)—Lateral border of the nasal aperture located directly underneath soft-tissue nasal ala furrow (naf’). 16. Subalare (sbal)—Inferior border of the nasal aperture located directly underneath soft-tissue subalare (sbal’). 17. Mediolateral philtrum (mlp)—Point located where a sagittal line passes through the midpoint between the widest point of the nasal aperture and the anterior nasal spine and intersects with an axial line passing midway through the upper jaw above the teeth and below the inferior nasal aperture. 18. Lateral philtrum (lp)—Point located where a sagittal line passes through the widest point of the nasal aperture and intersects with an axial line passing midway through the upper jaw above the teeth and below the inferior nasal aperture. 19. Zygion (zy)—The most lateral point of the zygomatic arch. 20. Supraglenoid (sgl)—Root of the zygomatic arch just before the ear. 21. Gonion (go)—Point located on the jaw line at the level of the angle between the posterior and the inferior borders of the mandible. 22. Supra M1 (sm1)—Point located on the alveolar process at the level of the middle of the first upper molar. 23. Occlusal line (ol)—Point located on anterior margin of the ramus of the mandible, in alignment with the plane of dental occlusion. 24. Infra M1 (im1)—Point located on the alveolar process at the level of the middle of the first lower molar.
Skeletal Landmarks
Midline points 1. Supraglabella (sg’)—The midline soft-tissue point directly overlying hard-tissue supraglabella (sg). 2. Glabella (g’)—The midline soft-tissue point directly overlying hard-tissue glabella (g). 3. Nasion (n’)—The midline soft-tissue point directly overlying hard-tissue nasion (n). 4. Rhinion (rhi’)—The midline soft-tissue point directly overlying hard-tissue rhinion (rhi). 5. Mid-philtrum (mp’)—The midline soft-tissue point midway between subnasale and the vermilion border of the upper lip. 6. Labrale superius (ls’)—The midline soft-tissue point at labiale superius on the vermilion border of the upper lip. 7. Labrale inferius (li’)—The midline soft-tissue point at labiale inferius on the vermilion border of the lower lip. 8. Mentolabial Sulcus (mls’)—The midline soft-tissue point directly overlying hard-tissue supramentale (sm). 9. Pogonion (pg’)—The midline soft-tissue point on the eminence of the soft-tissue chin overlying hard-tissue pogonion (pg). 10. Menton (m’)—The midline soft-tissue point directly overlying hard-tissue menton (m). Bilateral points 11. Frontal eminence (fe’)—The soft-tissue point directly overlying hard-tissue frontal eminence (fe). 12. Mid-Supraorbital (mso’)—The soft-tissue point directly overlying hard-tissue supraorbital (so). 13. Mid-Infraorbital (mio’)—The soft-tissue point directly overlying hard-tissue infraorbital (io). 14. Lateral orbit (lo’)—The soft-tissue point directly overlying hard-tissue lateral orbit (lo). 15. Nasal Ala Furrow (naf’)—Superior insertion of soft-tissue nasal ala furrow into lateral nose. 16. Subalare (sbal’)—The point at the lower limit of each alar base, where the alar base disappears into the skin of the upper lip. 17. Mediolateral philtrum (mlp’)—The soft-tissue point directly overlying hard-tissue mediolateral philtrum (mlp). 18. Lateral philtrum (lp)—The soft-tissue point directly overlying hard-tissue lateral philtrum (lp). 19. Zygion (zy’)—The soft-tissue point directly overlying hard-tissue zygomatic arch (za). 20. Supraglenoid (sgl’)—The soft-tissue point directly overlying hard-tissue supraglenoid (sg). 21. Gonion (go’)—The soft-tissue point directly overlying hard-tissue gonion (go). 22. Supra M1 (sm1’)—The soft-tissue point directly overlying hard-tissue supra m1 (sm1). 23. Occlusal line (ol’)—The soft-tissue point directly overlying hard-tissue occlusal line (ol). 24. Infra M1 (im1’)—The soft-tissue point directly overlying hard-tissue infra m1 (im1).
Corresponding Soft-Tissue Landmarks
TABLE 2––Skeletal- and soft-tissue anatomical landmarks used to measure facial soft-tissue depths (FSTDs) from cone beam computed tomography (CBCT) images.
276 JOURNAL OF FORENSIC SCIENCES
STARBUCK ET AL.
.
FACIAL TISSUE DEPTHS IN CHILDREN
277
FIG. 2––Anatomical landmarks located on cone beam computed tomography (CBCT) images to measure facial soft-tissue depths (FSTDs). Dolphin software was used to automatically calculate tissue depths between the hard and soft-tissue anatomical landmarks (shown overlaid upon each other here). Hard and soft-tissue landmark definitions are provided in table 2.
TABLE 3––Technical error of measurement (TEM) across each permutation of the three measurement trials. Midline FSTDs (mm) TEM trial 1 versus trial 2 TEM trial 2 versus trial 3 TEM trial 1 versus trial 3 Mean TEM Lateral FSTDs (mm) TEM trial 1 versus trial 2 TEM trial 2 versus trial 3 TEM trial 1 versus trial 3 Mean TEM
sg–sg’
g–g’
n–n’
rhi–rhi’
mp–mp’
ls–ls’
li–li’
mls–mls’
pg–pg’
m–m’
0.13
0.15
0.10
0.13
0.19
0.24
0.32
0.23
0.21
0.23
0.13
0.09
0.14
0.11
0.21
0.30
0.11
0.21
0.19
0.18
0.19
0.10
0.11
0.08
0.15
0.16
0.30
0.27
0.09
0.11
0.15 fe–fe’
0.12 mio– mio’ 0.22
0.11 lo–lo’
0.19 naf–naf’
0.23 sbal–sbal’
0.25 mlp–mlp’
0.24 lp–lp’
0.16 zy–zy’
0.17 sgl–sgl’
go–go’
0.22
0.11 mso– mso’ 0.17
0.25
0.17
0.21
0.12
0.13
0.14
0.22
0.20
0.13
0.15
0.31
0.19
0.16
0.12
0.14
0.12
0.18
0.17
0.20
0.19
0.14
0.19
0.11
0.12
0.20
0.16
0.19
0.25
0.17
0.18
0.12
0.13
ol–ol’
0.19
sm1– sm1’ 0.21
0.24
im1– im1’ 0.37
0.28
0.19
0.29
0.17
0.24
0.10
0.18
0.17
0.29
0.32
0.35
0.12
0.23
0.18
0.27
0.24
0.32
Overall Mean TEM = 0.19 mm.
TABLE 4––Midline mean facial soft-tissue depth (FSTD) values (mm) for unilateral and bilateral cleft lip and palate (CLP) individuals. Midline Tissue Depths Tissue Depth Measure (mm)
sg–sg’
Unilateral CL/CP (n = 35), Ages 7–11 Average 3.9 Standard Deviation 0.8 Median 3.9 Bilateral CL/CP (n = 19), Ages 7–11 Average 4.4 Standard Deviation 1.2 Median 4.0 Unilateral CL/CP (n = 20), Ages 12–17 Average 5.4 Standard Deviation 1.1 Median 5.6 Bilateral CL/CP (n = 12), Ages 12–17 Average 4.9 Standard Deviation 0.9 Median 4.8
g–g’
n–n’
rhi–rhi’
mp–mp’
ls–ls’
li–li’
mls–mls’
pg–pg’
m–m’
4.9 0.9 4.8
6.2 1.2 6.2
3.4 0.9 3.2
10.9 2.3 10.7
11.7 2.4 11.1
10.9 2.5 10.6
10.4 2.0 10.1
8.9 2.2 9.0
6.9 2.4 6.9
5.6 1.4 5.3
6.7 1.4 6.6
3.4 0.8 3.1
10.7 2.0 10.4
10.7 1.7 10.4
11.9 3.4 12.0
10.6 2.5 10.9
9.4 2.4 8.4
7.2 1.9 6.9
6.1 1.1 6.2
7.3 1.5 7.3
4.4 1.4 4.3
12.1 2.2 12.6
14.3 2.4 14.0
12.3 3.0 11.8
12.1 1.9 12.1
10.8 2.1 10.8
8.2 2.6 7.6
5.9 0.9 6.1
7.1 1.0 7.2
3.9 1.0 3.8
12.2 2.6 12.8
14.2 3.6 14.5
14.8 2.9 15.7
12.9 1.8 12.9
11.0 2.5 11.2
8.0 2.3 7.9
278
JOURNAL OF FORENSIC SCIENCES TABLE 5––Lateral mean facial soft-tissue depth (FSTD) values (mm) for unilateral and bilateral cleft lip and palate (CLP) individuals.
Bilateral Tissue Depths Tissue Depth Measure (mm)
fe–fe’
mso–mso’
mio–mio’
Unilateral CL/CP (n = 35), Ages 7–11 Average 4.4 6.1 6.7 Standard 1.1 1.4 1.7 Deviation Median 4.3 6.1 6.4 Bilateral CL/CP (n = 19), Ages 7–11 Average 4.9 6.9 7.5 Standard 1.0 2.0 2.0 Deviation Median 4.8 6.5 7.2 Unilateral CL/CP (n = 20), Ages 12–17 Average 5.5 8.3 7.5 Standard 1.2 2.8 1.7 Deviation Median 5.6 7.9 7.9 Bilateral CL/CP (n = 12), Ages 12–17 Average 5.9 8.1 8.0 Standard 1.5 1.5 2.6 Deviation Median 5.5 8.0 8.0
lo–lo’
naf–naf’
sbal–sbal’
mlp–mlp’
lp–lp’
zy–zy’
sgl–sgl’
go–go’
sm1–sm1’
ol–ol’
im1–im1’
12.4 2.7
13.5 2.5
12.4 2.3
12.1 2.7
12.6 2.8
8.1 1.9
10.4 2.4
11.4 3.2
28.4 3.8
20.2 2.7
22.7 3.3
12.4
13.2
12.4
12.0
12.4
7.8
10.0
11.0
29.1
19.8
22.9
13.9 2.8
14.5 2.1
14.8 2.6
11.1 2.6
14.9 3.1
9.5 2.7
10.1 1.6
13.2 5.1
29.4 3.9
20.7 3.7
23.7 4.4
13.6
14.6
14.5
11.1
14.9
8.8
9.9
12.2
27.9
20.0
22.3
13.9 3.2
18.3 7.8
14.8 2.9
13.6 2.7
14.9 2.7
9.4 2.5
11.8 3.1
13.7 4.7
30.1 5.1
21.7 3.8
24.5 4.9
14.4
17.0
15.0
13.8
14.4
8.7
11.5
13.0
31.0
21.6
24.7
13.4 4.1
16.8 2.8
15.5 2.6
13.5 3.1
15.9 2.2
9.4 3.0
11.0 2.2
14.7 6.6
33.6 7.7
24.4 6.7
27.4 7.0
14.0
17.7
15.6
12.3
16.5
9.4
11.0
12.1
33.6
22.0
25.3
ity across three repeatability trials. The ICC value was 0.99 and indicated that FSTD measurements were strongly reliable. The technical error of measurement (TEM) was calculated for each permutation of measurement error trials to assess intra-observer error using the following: rP ffiffiffiffiffiffiffiffiffiffiffiffi D2 ; TEM ¼ 2N
FIG. 3––Mean facial soft-tissue depth (FSTD) comparison (mm) among age groups born with the same cleft lip and palate (CLP) phenotype. (A) FSTDs among the two unilateral CLP age groups (dashed and solid line). (B) FSTDs among the two bilateral CLP age groups (dashed and solid line).
=RAND()*100) and sorting the randomly generated numbers and coded image column numbers. The intraclass correlation (ICC) coefficient was calculated to assess intra-observer reliabil-
where D was equal to the difference of measured values between two trials, and N was equal to the number of individuals measured (17). TEM values for each FSTD measurement comparison are provided (Table 3) with the overall mean TEM of 0.19 mm being considered adequately low for the purposes of this investigation. As previous investigations have noted (37,56), it is especially important to minimize measurement error from imaging artifacts and human error. Individuals in this study were only imaged once to minimize radiation exposure to living subjects, thereby precluding the assessment of measurement error associated with the CBCT i-Cat machine. As a result, it is possible that errors are larger than those reported. After assessing reliability and TEM, FSTDs (Fig. 2) were measured on each image on three separate occasions with at least 24 hours between each measurement session. FSTD measurements that differed by more than 1 mm from homologous FSTD values collected from the same individual during different trials were remeasured and replaced to minimize measurement error occurring from human error and imaging artifacts. Individual FSTD values for each age group were tested for normal distributions using a Kolmogorov–Smirnov test for normality (a ≤ 0.05). Scatterplots and regression lines were explored for FSTD values and age. Average FSTD values and previously published pooled FSTD values for children (37) were graphed to explore similarities and differences. Data were analyzed using a multivariate analysis of variance (MANOVA)
STARBUCK ET AL.
.
FACIAL TISSUE DEPTHS IN CHILDREN
279
FIG. 4––Scatterplots and regression lines for facial soft-tissue depths (FSTDs) (y-axis in mm) and age (x-axis in years) combined for unilateral and bilateral cleft lip and palate (CL/CP) samples.
followed by separate post hoc ANOVA tests for independent variables showing significant effects on the dependent variables. MANOVA is a statistical test for assessing the null hypothesis of the equality of multiple averages and is used to determine whether one or more independent variables have a significant effect on two or more dependent variables. MANOVA is appropriate here because FSTDs represent multivariate data measured from the facial complex, and this method is capable of detecting statistical differences that may not be discovered when only using univariate statistics. The independent variables used in the MANOVA analysis are cleft type (unilateral/bilateral), age (7– 17 years), age group (younger 7–11 years/older 12–17 years), sex (male/female), and bone graft status (yes/no), and the dependent variables are all of the unilateral and averaged bilateral FSTD values. Post hoc ANOVA tests were carried out to determine which FSTDs were significantly affected by independent
variables shown to be significant by the MANOVA. Statistical significance was initially set at p-value ≤ 0.05, but lowered to p-value ≤ 0.016 after a Bonferonni adjustment for 3 tests (one MANOVA and two ANOVAs). Results Summary statistics for midline and lateral tissue depth values are provided in Tables 4 and 5. Average FSTD values for the unilateral and bilateral CL/CP age groups (i.e., 7–11 years and 12–17 years) are equivalent, and patterns of variation for the two age groups across the face are nearly identical to each other (Fig. 3A and B). Average FSTD values for younger (7–11 years) and older (12–17 years) age groups follow similar patterns of variation, although FSTDs in the older age group tend to be thicker (Fig. 3A and B). Scatterplots of tissue depths versus age
280
JOURNAL OF FORENSIC SCIENCES
FIG. 5––Scatterplots and regression lines for facial soft-tissue depths (FSTDs) (y-axis in mm) and age (x-axis in years) combined for unilateral and bilateral cleft lip and palate (CL/CP) samples (continued).
with fitted regression lines were created (Figs 4 and 5) and show that all FSTDs have positive slopes and tend to increase with age on average. A MANOVA revealed a significant multivariate main effect for cleft type (p-value = 0.034) and age (p-value
