Facial height in Japanese-Brazilian descendants with normal occlusion

OBJECTIVE: The aim of this study was to determine the standards of facial height in 30 young (14-year-old) Japanese-Brazilian descendants with normal occlusion, and assess whether sexual dimorphism is evident. METHODS: The cephalometric measurements used followed the analyses by Wylie-Johnson, Siriwat-Jarabak, Gebeck, Merrifield and Horn. RESULTS: Results showed dimorphism for total anterior facial height (TAFH), lower anterior facial height (LAFH), anterior facial height (AFH), total posterior facial height (TPFH) and upper posterior facial height (UPFH) measurements. CONCLUSIONS: The standards of facial heights in young Japanese-Brazilian descendants with normal occlusion were observed. Sexual dimorphism was identified in five out of thirteen evaluated variables at this age range.


INTRODUCTION
Within the context of contemporary Orthodontics, making accurate diagnosis and prognosis determines whether a clinician can provide patients with the best cost-benefit treatment. Based on principles of effectiveness and efficiency, only one or two treatment alternatives best fit patient's esthetic, functional and psychological needs. 1 Therefore, clinicians must use all possible resources to achieve this ideal goal on orthodontic practice.
Cephalometry is a valuable auxiliary diagnostic tool as it allows the relationship among bone structures, dental tissue and soft tissue to be determined by means of lateral radiographs, 6 thereby facilitating complete assessment of malocclusion in different space dimensions, including anterior-posterior 25 and vertical. 30 Analyzing malocclusion this way allows understanding of how and in what direction and manner each element of the stomatognathic system contributes to its conformation. For this reason,the use of cephalometric analysis is rendered necessary in the anterior-posterior and vertical directions, and so it is to analyze the influence of vertical changes in the severity of malocclusion in anterior-posterior direction. 23 Vertical facial changes influence mandibular position and rotation, either clockwise or counterclockwise, thereby contributing to the development of deep or open bite. Thus, orthodontic treatment should induce desirable changes and minimize inevitable undesirable ones. 22 To assess vertical facial changes, new cephalometric analyses were developed, 24 and the present study uses measurements employed by Wylie and Johnson, 30 Siriwat and Jarabak, 24 Gebeck, 8 Merrifield 17 as well as Horn,9 all of which assess anterior and posterior facial height, facial ratios and facial height index, measurements which were used in previous studies. 5,26 In these studies, cephalometric standards from different racial and ethnic groups and with miscegenation were determined and compared, showing the need for individualization, which has also been reported in worldwide literature. 5,7,13,18,26 The studies found in the literature were devoted to certain groups, such as Caucasians or Mongoloids, but not to the result of their miscegenation. For this reason, explaining the need for individualization and understanding of cephalometric characteristics of different miscegenation patterns is important.
To this end, the following were assessed: Specific cephalometric patterns of anterior and posterior facial heights; facial ratios and facial height index 9 for young Japanese-Brazilian descendants with normal occlusion using variables in the vertical direction of the face. The presence of sexual dimorphism was also assessed.

MATERIAL AND METHODS
A total of 30 lateral cephalometric radiographs of young Japanese-Brazilian descendants (15 males and 15 females with an average age of 14 years) with normal occlusion, selected from elementary and high schools located in the city of Bauru, were assessed. The selected patients had the following characteristics: Japanese-Brazilian descendent of parents and/or grandparents from Japan, except for the island of Okinawa, and Caucasian Brazilian parents (Portuguese, Spanish or Italian ancestry); aged between 11.91 to 16.61 years; with normal occlusion; and no history of previous orthodontic treatment.
All patients had permanent teeth in occlusion, except for third molars. Additionally, they had normal molar relationship, mild or absent crowding, no crossbite, normal overbite and overjet, no differences between mandibular positions in centric relation and maximum intercuspation, and well-balanced faces.
Cephalometric radiographs were obtained according to the standards recommended by the Department of Radiology, School of Dentistry/University of São Paulo, Bauru. Radiographic image magnification (using a Siemens equipment) was of 9.8%, corrected during measurements of radiographs so as to increase accuracy of the method employed.

Preparation of cephalograms
Anatomical tracing was carried out according to the recommendations described by Interlandi 11 and Vion; 28 the average of anatomical structures was used when two radiographic images of the same structure were identified. The following anatomical structures were assessed (Fig 1): Sella turcica, clivus, external cortex of the frontal bone and nasal bones; mean of pterygomaxillary fissure; mean of inferior borders of orbits; average of external auditory meatus; maxilla, mandible, teeth (upper and lower central incisors and first molars) and soft tissue profile. After performing the anatomical tracing, landmarks were identified and then digitized by the digitizing tablet AccuGrid XNT, model A30TL.F (Numonics Corporation, Montgomeryville, PA, USA). Data were processed using Dentofacial Planner Software, version 7.02 (Dentofacial Planner Software Inc.,Toronto, Ontario, Canada) installed in a PC with 700MHz Intel Pentium III processor.

Cephalometric landmarks, lines and planes
After the anatomical tracing was prepared, cephalometric landmarks were located according to Miyashita: 19 S (Sella), N (Nasion), ANS (Anterior Nasal Spine), PNS (Posterior Nasal Spine), Me (Menton), Go (Gonion) and Ar (Articulare) (Fig 1). After locating the cephalometric landmarks that are independent of guidance tracing, plans and lines were drawn, and Ar' and ANS' points were constructed according to Wylie and Johnson 30 as well as Siriwat and Jarabak 24 (Fig 1).

Cephalometric measurements in vertical direction
The measures taken according to the analysis by Wylie-Johnson 30 are shown in Figure 2 The measures used according to the Siriwat and Jarabak 24 analysis are shown in Figure 3.

STATISTICAL METHOD Descriptive and comparative analyses
Means and standard deviation were used to describe the sample of Japanese-Brazilian descendants. To investigate the existence of sexual dimorphism, t-test with significance level set at 0.05 was applied due to normal distribution of variables in the Kolmogorov-Smirnov test. All statistical analyses were performed using Statistica software (Statistica for Windows 6.0, Statsoft, Tulsa, OK).

Method error
Cephalometric tracings and measurements of 50% of the sample were remade by the same examiner a month after obtaining the initial cephalograms.
Systematic and casual errors were independently assessed for each cephalometric variable, as recommended by Houston. 10 Systematic error was calculated by dependent t-test for paired samples. Casual error was calculated by Dahlberg's formula 4 using the standard deviation of differences between repetitions.

RESULTS
Results are divided and presented in tables for didactics purposes and to favor visualization and understanding.

CASUAL AND SYSTEMATIC ERROR
Casual error was determined by Dahlberg's formula, 4 whereas systematic error was assessed by dependent t-test. 10 Statistical analysis carried out to assess intra-examiner error revealed no systematic errors. Casual errors, however, were minimal, since measurements were linear and most of them had a value lower than 1 mm. Only two variables, TAFH and UAFH, yielded slightly higher causal error values: 1.44 mm and 1.25 mm, respectively.
Facial height in Japanese-Brazilian descents with normal occlusion original article

CHARACTERIZATION OF THE SAMPLE
The sample comprised 30 Japanese-Brazilian descendants, 15 males and 15 females, with mean age of 14 years old -14.78 years for males and 13.22 years for females, representing an age difference of 1.56 years of which significance was tested and confirmed by independent t-test set at 5% significance level. Characterization of vertical facial growth pattern by means of SN.GoGn variable showed an average of 33.08 degrees (33.02 for males and 33.15 for females), with no statistically significant difference.

SAMPLE COMPARATIVE AND DESCRIPTIVE ANALYSES
Descriptive analysis determined the number of research subjects, means, standard deviations as well as minimum and maximum values of the population necessary for a confidence interval of 95% for each variable considered in the current study (Table 1).
Sexual dimorphism was assessed by means of independent t-test set at 5% significance level. Average male and female data with respective standard deviations and P-values are presented in Table 2.

DISCUSSION
Facial vertical pattern affects facial harmony and attractiveness. In this context, orthodontic treatment can favor or disfavor balance by implementing facial changes in the vertical direction of which even lay people are aware of. 21 Therefore, clinicians should have an individualized reference 20 to conduct orthodontic treatment in order to induce the desired changes and minimize undesirable, inevitable ones. 2 This study should be viewed as primarily descriptive. It aims at demonstrating how the values of young Japanese-Brazilian descendants are incomparable to values previously established for Caucasian and Mongoloid subjects. It also aims at analyzing sexual dimorphism for each variable. Thus, the values determined for the variables analyzed herein should be compared to other results previously reported in the literature with a view to further investigate this topic.
This discussion of results is divided into anterior facial height and its ratios, posterior facial height and its ratios, and Facial Height Index (Horn 9 ). Each of these sections was divided into sub-sections so as to favor interpretation of results.

ANTERIOR FACIAL HEIGHTS TAFH -Total anterior facial height
TAFH for Japanese-Brazilian descendants had an average value of 125.90 mm for males and 119.74 mm for females. Statistically significant difference, with significance level set at 5%, was identified between these values, thereby indicating sexual dimorphism with greater vertical development for males.  This may have been caused by age difference between males and females. However, there is a chronological gap between growth and development of males and females in the phase of adolescence, including the vertical development of the face. Additionally, females in general have their pubertal growth spurt at an earlier age than males. Therefore, growth will likely be more balanced between males and females in this condition, with a mean age difference of 1.56 years during adolescence, particularly because females represented the group with the lower average age. Similar findings were also reported in other studies. 14,26 The values determined for Japanese-Brazilian descendants are close to the highest values found in the literature for Caucasians, but were even closer to values found for Mongoloids. Ishii et al 12 conducted a study in which significant differences were found between Japanese Mongoloid and British Caucasian groups for both males and females, with the Mongoloid group presenting the highest values. Takahashi 26 also found significant differences between Caucasian and Mongoloid racial groups, particularly for males, with the largest values found in the Mongoloid group. However, for females, no significant differences were found among racial groups. Additionally, the female Mongoloid group had higher values of TAFH.
Although the values found in the literature showed great variability for the TAFH variable, in general, the values reported in this study were very close to those found in the literature for Japanese 12 and their descendents 26 within a similar age range. Disagreement among some values found in the literature 12,14,23,24,26 explains the large variation among them (Fig 5).

UAFH -Upper anterior facial height
Young Japanese-Brazilian descendants showed an average UAFH value of 52.52 mm (53.40 mm for males and 51.64 mm for females) with no statistically significant difference at 0.05 significance level between them. Thus, sexual dimorphism was not evident, thereby implying that upper facial height does not contribute to dimorphism found in TAFH.
Results showed no differences between males and females, confirming the findings by Domiti et al 5  The values for young Japanese-Brazilian descendants are between those found in the literature for Caucasians, 6,27 but once more are closer to those reported by Takahashi 26 for the Mongoloid group.
Ishii et al 13 showed that the upper anterior facial height was significantly higher in the Japanese Mongoloid group in comparison to the British Caucasian group. Takahashi 26 found a significant difference comparing Caucasian and Mongoloid racial groups for males, but not for females. He also observed higher values for the Mongoloid group when comparing males and females of both races. Figure 6 shows the values found in the literature for UAFH. 5,14,23,26,27,29

LAFH -Lower anterior facial height
The mean LAFH values for young Japanese-Brazilian descendants are 70.29 mm, 72.49 mm for males and 68.1 mm for females. Values were statistically different for males and females, thereby featuring sexual dimorphism and confirming the findings by other authors such as Lock 15 and Miyajima. 18 However, Domiti et al 5 and Takahashi 26 found no differences between males and females for either one of the two racial groups. Ursi et al 27 identified differences between males and females older than 16 years  with LAFH values higher for males at this age. Sexual dimorphism in the Japanese-Brazilian descendent sample leads us to the conclusion that lower anterior facial height contributed significantly to the dimorphism found in TAFH.
The values determined for the young Japanese-Brazilian descendants are closely related to the highest values found in the literature for Caucasians 6 and Mongoloids. 18 Ishii et al 13 reported that Japanese Mongoloid individuals have LAFH values significantly higher than British Caucasian individuals, although Takahashi 26 found significant differences between Caucasian and Mongoloid racial groups, only for males, thereby demonstrating greater LAFH values in the Mongoloid group for both males and females. Figure 7 shows the values found in the literature 5,18,23,26,27,29 with a large variation for LAFH values.

UAFH/TAFH ratio
Young Japanese-Brazilian descendants showed an average UAFH/TAFH ratio of 42.76%, being 43.10% for females and 42.42% for males, with no statistically significant difference between these values, at 0.05 significance level. Consequently, no sexual dimorphism was evident, corroborating the findings by Wylie and Johnson, 30 as well as Takahashi 26 -who did not find a statistically significant difference between males and females for both study groups at 0.05 significance level.
The values determined for young Japanese-Brazilian descendants are closer to the minimum value found by Locks 16 for Caucasians (42%).
Takahashi 26 also found no significant differences when comparing Caucasian and Mongoloid males and females, thus showing a balance in this ratio. Figure 8 shows the comparison among values found in the literature. 3,14,16,26,30 LAFH/TAFH ratio Young Japanese-Brazilian descendents showed an average UAFH/TAFH ratio of 57.23%, being 56.89% for females and 57.57% for males, with no statistically significant difference between these values, at a 0.05 significance level. Consequently, no sexual dimorphism was evident, corroborating the findings by Takahashi 26 who found no statistically significant difference between males and females for both study groups.
The values determined for young Japanese-Brazilian descendants are close to the highest values found in the literature for Caucasians. 16, 23 Takahashi 26 also found no significant differences when comparing Caucasian and Mongoloid males and females, thus showing a balance in this ratio. Figure 9 shows the values found in the literature for LAFH/TAFH. 3

POSTERIOR FACIAL HEIGHTS TPFH -Total posterior facial height
Young Japanese-Brazilian descendants showed an average TPFH of 81.60 mm, being 84.36 mm for males and 78.85 mm for females, with a statistically significant difference at a 0.05 level, thereby indicating sexual dimorphism with greater development of male posterior facial height. As discussed regarding TAFH dimorphism, it can be inferred that this is not a simple reflection of age difference between males and females. In addition, a similar condition was reported by Takahashi 26 for the Mongoloid group, although Chang et al 3 did not find this difference.
The values determined for young Japanese-Brazilian descendants are closer to the values found by Takahashi 26 for the Mongoloid group. This author also found significant differences when comparing Caucasian and Mongoloid males and females, and reported that the Mongoloid group had higher values 26 .
The values reported in the literature for TPFH 23,24,26 are shown in Figure 10.

UPFH -Upper posterior facial height
The mean UPFH value of young Japanese-Brazilian descendants was 34.47 mm, being 36.24 mm for males and 32.70 mm for females. Values were statistically different for males and females with significance level set at 0.05. Thus, sexual dimorphism Takahashi 26 also identified significant differences when comparing Caucasian and Mongoloid racial groups, for males and females, with higher values for Mongoloids. These findings differed from those by Ishii et al 13 who found no difference between Japanese Mongoloid and British Caucasian.
Comparison between values found in this study and by Takahashi 26 indicate greater proximity between the values of young Japanese-Brazilian descendants and Mongoloids, with higher values for the first group, as presented in Figure 11. This finding can neither be attributed to differences in methodology, which was the same, nor to mean age difference, since the mean age of the Mongoloid group was greater (15.71 years) in Takahashi's study 26 . However, this finding may be due to the use of a different sample, with a slightly more vertical pattern of young Japanese-Brazilian descendants, or because of race miscegenation that generates a new biological and genetic conformation.

LPFH -Lower posterior facial height
Young Japanese-Brazilian descendants showed an average LPFH value of 47.13 mm, being 48.12 mm for males and 46.14 mm for females. No sexual dimorphism was evident, thereby corroborating the results by Takahashi 26 f r both groups.
Lack of dimorphism in LPFH values of young Japanese-Brazilian descendants inferred that LPFH does not contribute to the dimorphism found in TPFH.
Takahashi 26 identified significant differences when comparing Mongoloid and Caucasian males and females, with the Mongoloid group showing higher values. The values of young Japanese-Brazilian descendants are closer to the maximum values obtained with Caucasians, 23 and even closer to the values reported for Mongoloids. 12,26 This condition is well characterized in F igure 12.

UPFH/TPFH ratio
Young Japanese-Brazilian descendants showed an average UPFH/TPFH ratio of 42.23%, being 42.98% for males and 41.48% for females, with no statistically significant difference at a 0.05 significance level. Therefore, no sexual dimorphism was observed. This result corroborates the findings by Takahashi 26 for the Mongoloid group, although this author reported sexual dimorphism with higher UPFH/TPFH ratios for males in the Caucasian group.
Takahashi 26 found no significant difference when comparing Caucasian and Mongoloid males and females. The values for young Japanese-Brazilian descendants are close to those reported by Takahashi,26 as shown in Figure 13, despite age difference and the use of a different sample with its own racial miscegenation. The cause may be stability of values for this variable after a certain age and a small variation between different races and their miscegenations.

LPFH/TPFH ratio
Young Japanese-Brazilian descendants showed an average LPFH/TPFH value of 57.77%, being 57.02% for males and 58.51% for females. The lack of dimorphism in these results corroborates Takahashi 26 for the Mongoloid group, although this author reported sexual dimorphism in the Caucasian group with higher LPFH/TPFH values for females. Takahashi 26 found no significant differences when comparing Mongoloid and Caucasian males and females. The values for young Japanese-Brazilian descendants are close to those reported by Takahashi 26 for the two groups, as shown in Figure 14.

Determining the facial height index (FHI)
Horn 9 proposed a variable to track patient's vertical dimension during treatment. The proposed index is calculated by dividing the posterior facial height (PFH, the distance in millimeters from point   Ar to the mandibular plane) and the anterior facial height (AFH) (the distance in millimeters from the palatal plane to the point Me). According to the author, the use of the facial height index 9 is an additional aid in the diagnosis of excess or reduced vertical dimension, allowing observation of vertical dimension during treatment and adjustment of orthodontic mechanics to offset any unfavorable trend.   The values determined for young Japanese-Brazilian descendants are closer to the maximum values found in the literature for Caucasians, 9,17 and closer to the values reported by Takahashi 26 for the Mongoloid group. The variation in AFH found in the literature is shown in Figure 15.  Takahashi 26 identified significant differences when comparing Caucasian and Mongoloid males, and found no significant differences among racial groups for females. The same author also reported that, with regard to both males and females, Mongoloids had higher values than Caucasians.

PFH -Posterior facial height
The mean PFH value of young Japanese-Brazilian descendants was 50.33 mm, being 51.16 mm for males and 49.51 mm for females. No sexual dimorphism was found with these results, similar to what was observed in Takahashi's 26 study for both Mongoloid and Caucasian groups.
The values determined for young Japanese-Brazilian descendants are close to those found in the literature for Caucasians 17 and even closer to those reported for the Mongoloid group. 26 Takahashi 26 also reported significant difference when comparing Caucasian and Mongoloid males and females. The variation in PFH values found in the literature 9,17 is shown in Figure 16.

FHI -Facial Height Index (Horn 9 )
The mean FHI value for Japanese-Brazilian descendants (Horn 9 ) was 72.65%, being 71.82% for males and 73.48% for females. No sexual dimorphism was found, thereby corroborating the results by Takahashi 26 for the Mongoloid group, although this author reported sexual dimorphism in the Caucasian group.
A small variation in FHI values was observed in the literature, 9,23,26 as shown in Figure 17. For this variable, the results of the present study were similar to those reported by Takahashi 26 for Caucasians and Mongoloids. He also reported significant differences between Caucasian males, but not for females.
Results for young Japanese-Brazilian descendants showed, in general, that the values of variables and the analysis of sexual dimorphism -particularly when compared with those reported by Takahashi 26 -suggested a closer relationship between the values of Japanese-Brazilian descendants and the Mongoloid group for all variables except for the ratios UAFH / TAFH, LAFH / TAFH, UPFH / TPFH, and LPFH / TPFH as well as FHI. Thus, further comparison between young Japanese-Brazilian descendants, Caucasians and Mongoloids should be performed in a study with the specific objective of precisely establishing the relationship between these groups.
Moreover, the analysis of sexual dimorphism of young Japanese-Brazilian descendants suggests a more vertical pattern for males, characterized by higher values in the variables TAFH, LAFH, AFH, TPFH and UPFH. However, the variables denoting vertical facial ratio do not show the same trend.

CLINICAL CONSIDERATIONS
Vertical facial changes influence mandibular position and rotation, either clockwise or counterclockwise, 22,23 thereby contributing to the development of deep bite or open bite, and potentially increasing the severity of anterior-posterior malocclusion. Thus, orthodontic treatment should induce desirable alterations and minimize the undesirable ones when the latter are inevitable. 2 Therefore, malocclusion should be analyzed completely and in all different dimensions of space so as to favor understanding of how and in what direction each element of the stomatognathic system contributes to the conformation of malocclusion, which is necessary for cephalometric analysis of anterior-posterior and vertical directions, as well as analysis of the influence of vertical changes in the severity of antero-posterior malocclusion. 23 However, the literature indicates that malocclusion analysis requires full assessment so as to individualize cephalometric norms regarding patient's sex, age and race. 20 Thus, cephalometric standards from different ethnic and racial groups and miscegenations were determined and compared, and the need for individualization for each specific group was demonstrated 5,7,18,20,26 to better understand and assess the cephalometric characteristics of different groups and miscegenations with respect to orthodontic diagnosis and planning. Thus, the present study provides the clinician with a more specific reference in the vertical direction of the face, particularly for young Japanese-Brazilian descendants with normal occlusion.
Moreover, as a topic for future research, the values of Japanese-Brazilian descendants should be compared with those of other subjects, particularly Mongoloid and Caucasian Brazilians.

CONCLUSIONS
Based on the sample and methods employed herein, values are presented to establish a cephalometric pattern of anterior and posterior facial heights and its ratios, as well as the facial height index (Horn 9 ) for young Japanese-Brazilian descendants with normal occlusion. Results revealed the presence of sexual dimorphism in the following cephalometric measurements: TAFH, LAFH, AFH, TPFH and UPFH.