Dentofacial biometry as a discriminant factor in the identification of remote Amazon indigenous populations

Introduction

This study aimed to examine the role of dentofacial morphology in discriminating semi-isolated indigenous groups. These populations present a similar pattern of dietary habits. Studies in human genetics have reported a large intertribal genetic distance and low intratribal variation.

Methods

This study was conducted following the Strengthening the Reporting of Observational Studies in Epidemiology guidelines. Face dimensions were measured through facial photogrammetry, and dental arches and tooth size were evaluated using plaster models. A total of 98 subjects in the permanent dentition and belonging to 4 indigenous groups were analyzed: Arara-Iriri (n = 20), Arara-Laranjal (n = 33), Assurini (n = 25), and Xicrin-Kayapó (n = 20). The random and systematic errors were verified using the Dahlberg formula and the intraclass correlation coefficient (ICC), respectively. In order to evaluate the discrimination of the variables to identify the indigenous groups, a discriminant analysis was performed ( P <0.05).

Results

A small causal error (Dahlberg, 0.13-1.81) and excellent replicability (ICC, 0.75-0.99) for face dimensions as well as for models (Dahlberg, 0.10-0.68; ICC, 0.94-0.99) were observed. The discriminant analysis allowed the identification of 4 populations by facial and dental arch dimensions and tooth size. Dentofacial biometry revealed an accuracy of 98% for females and 100% for males, which confirms a high intragroup homogeneity and considerable intergroup heterogeneity for dentofacial features.

Conclusions

Biometric measurements of the human face added with tooth size and dental arch dimensions are very useful to identify remote indigenous populations with high accuracy. Supported by previous studies in human genetics, these findings reinforce the role of genetic variation in the determination of dentofacial features.

Highlights

  • We examined the ability of dentofacial morphology in discriminating indigenous populations.

  • Dentofacial biometry revealed an accuracy of 98% for females and 100% for males.

  • The identification of females was obtained primarily by the face dimensions.

  • Males were discriminated mainly by tooth size and dental arch dimensions.

  • Supported by previous studies, our findings reinforce the role of genetics in dentofacial features.

Forensic dentistry, a branch of dental anthropology and forensic medicine, plays an important role in the process of identifying an individual. The most relevant step is to construct the biological profile, which includes the possibility of determining sex, estimating age and ancestry, and identifying specific trauma and pathological lesions. , The possibility of determining specific morphological characteristics that are capable of discriminating individuals was demonstrated in studies examining face and skull morphology.

Face morphology has been used to discriminate ethnic groups in individuals. However, despite this ability, the accuracy of this method is still far from precise identification, even in remote populations. In urban populations, discrimination has a lower accuracy because of larger genetic miscegenation. This difference is explained by genetic drift, an ectostatic process that acts more frequently on small and isolated populations, inducing loss of genetic variation and the prevalence of specific characteristics in the population. ,

Beside face morphology, discriminant analysis has innumerable purposes in the evaluation of human dentition. One of its uses is the classification of individuals through dental data such as tooth wear and size, dental arch dimension, and dental crowding, because it is commonly accepted that patterns for skeletal identification vary between different populations, and there is the possible occurrence of changes in populations over time.

Determination of how the dentofacial morphological characteristics occurred among human isolated populations with similar diets and a considerable genetic distance, , , through biometric examination before the eventual miscegenation process, can allow the classification of individuals in each group if these characteristics are genetic.

A previous investigation has shown that, although deficient, there is a certain capacity of identification from the morphological analysis of the face when analyzed in isolation, and dental arch dimensions are predominantly associated with genetic variation among indigenous populations. The objective of this study was to evaluate the discrimination of Amazon indigenous groups, taking into account the analysis of face morphology added to tooth size and dental arch dimensions. These data will allow us to contribute to the understanding of the role of genetics and the environment in the determination of dentofacial characteristics.

Material and methods

This cross-sectional study was carried out in indigenous populations located in the Medium Valley of the Xingu River, State of Pará, Brazil ( Fig 1 ). The Strengthening the Reporting of Observational Studies in Epidemiology guidelines for cross-sectional studies were followed.

Fig 1
Map of South America. The magnified box highlights the Medium Valley of the Xingu River, Amazon, Brazil. Location of Arara-Iriri, Arara-Laranjal, Assurini, and Xicrin-Kayapó villages.

The indigenous population were informed of the objectives of the study and authorized their participation through the signature of the free and informed consent form, translated into their source language. The adults responsible for children and minors involved in the study were required to authorize the participation on their behalf. For adults who could not sign, verbal consent was obtained through audio recording. The study protocol and informed consent were approved by the Brazilian National Research Ethics Committee (CONEP), under the number 25000.066559/2010-11. The legal permission for researchers to enter indigenous territory was obtained from the National Indian Foundation after the favorable opinion of scientific merit obtained by the National Council for Scientific and Technological Development.

A total of 665 indigenous individuals from 4 Amazonian villages were evaluated, including Arara-Iriri (n = 80), Arara-Laranjal (n = 242), Assurini (n = 154), and Xicrin-Kayapó (n = 189). The inclusion criteria were indigenous in the permanent dentition and without dental loss except for third molars. A final sample of 98 participants (51 female, 47 male) from 4 villages (Arara-Iriri [n = 20], Arara-Laranjal [n = 33], Assurini [n = 25] and Xicrin-Kayapó [n = 20]) was included ( Table I ).

Table I
Characterization of sample size, sex, and mean age (minimum and maximum value) by indigenous villages
Sex Arara-Iriri (n = 20) Arara-Laranjal (n = 33) Assurini (n = 25) Xicrin-Kayapó (n = 20)
n (%) Mean age, y (min-max) n (%) Mean age, y (min-max) n (%) Mean age, y (min-max) n (%) Mean age, y (min-max)
Female 10 (50) 17.0 (11.4-47.4) 19 (57.6) 15.8 (11.7-23.1) 13 (52) 15.1 (10.6-20.9) 9 (45) 14.6 (11.6-19.5)
Male 10 (50) 20.7 (11.3-32.3) 14 (42.4) 20.6 (11.1-48.0) 12 (48) 17.0 (12.0-25.3) 11 (55) 14.5 (11.2-19.3)
Total 20 (100) 18.9 (11.3-47.4) 33 (100) 17.8 (11.1-48.0) 25 (100) 16.0 (10.6-25.3) 20 (100) 14.6 (11.2-19.5)

Intraoral and facial photographs were taken with a digital camera with 18-megapixel resolution (Rebel T3i; Canon, Tokyo, Japan). Dental casts were obtained from the maxillary and mandibular dental arches with an irreversible hydrocolloid.

Standardized frontal and lateral facial photographs were obtained from the face of each selected indigenous individual in the natural head position, obtained according to the method described by Scavone. A slight modification in obtaining the natural head position was required because of the difficulty faced by some indigenous subjects in looking directly at their face when in front of the mirror.

The photographs, obtained by a single operator, were exported to the 3.0 ImageTool program (University of Texas Health Science Center, San Antonio, Tex); 14 linear and angular measures in pixels were obtained and were subsequently converted to millimeters. The facial measures studied were prominence of the glabella relative to the true vertical line (LVV-Gb), with this line referenced for measurements of nose prominence (LVV-No), A and B point in soft tissue (LVV-A’ and LVV-B’), upper and lower lip protrusion (LVV-Ul and LVV-Ll), and chin prominence in soft tissue (LVV-Pog’). Vertical analysis of the face was obtained, including the height of the middle third (Gb-Sn), lower third (Sn-Me’), and total height (Gb-Me’), as well as the ratios between the height and width of the face (TFH/FW) and lower facial height and total facial height. The angular measures that were noted included profile angle (Gb.Sn.Pog’) and nasolabial angle (Cm.Sn.Ul).

The facial analyses used were those described by Arnett and Bergman. The identification of the facial soft tissue points and the cephalometric variables ( Fig 2 ) were obtained by 2 previously calibrated examiners, with the calibration being supported by the intraclass correlation coefficient (ICC) to verify the error of the method ( Table II ). The final measure was determined from the mean value of the 2 examiners.

Fig 2
Angular and linear measurements used in lateral and frontal facial photographs. Zy , zygomatic (Zyr’- right; Zye’-left); Gb , glabella; No , nasal apex; Cm , columella; Sn , subnasal; A’ , the A point in soft tissue; Ul , upper lip; Ll , lower lip; B’ , the B point in soft tissue; Pog’ , pogonium in soft tissue; Me’ , mentum in soft tissue.

Table II
Mean, SD, random error, and systematic error of facial analysis variables in each indigenous group
Variables (degrees or mm) Random error (Dahlberg) Systematic error ICC, (95% CI) Arara-Iriri (n = 20) Arara-Laranjal (n = 33) Assurini (n = 25) Xicrin-Kayapó (n = 20)
Mean SD Mean SD Mean SD Mean SD
LVV-Gb (mm) 0.28 0.98 (0.95, 0.99) –11.65 4.74 –9.28 3.71 –12.93 3.63 –11.44 2.88
LVV-No (mm) 0.15 0.98 (0.95, 0.99) 10.78 1.21 9.73 2.20 11.60 1.59 10.71 1.49
LVV-A’ (mm) 0.32 0.76 (0.48, 0.90) 1.24 1.03 1.16 1.05 0.41 1.53 2.06 0.97
LVV-Ul (mm) 0.15 0.99 (0.97, 0.99) 5.06 1.64 4.19 1.41 3.67 2.33 5.93 1.71
LVV-Ll (mm) 0.13 0.95 (0.88, 0.98) 2.56 2.58 2.07 2.16 1.97 2.78 2.86 2.03
LVV-B’ (mm) 0.15 0.94 (0.86, 0.98) –2.77 3.61 –3.42 2.88 –4.43 2.88 –4.40 2.85
LVV-Pog’ (mm) 0.23 0.75 (0.46, 0.89) –1.61 5.20 –2.70 3.45 –3.23 3.53 –4.08 3.40
Gb-Me’ (mm) 1.81 0.96 (0.90, 0.99) 129.21 9.39 118.25 10.27 135.41 10.36 135.41 10.36
Gb-Sn (mm) 0.99 0.94 (0.85, 0.98) 63.77 4.97 58.14 5.53 66.16 4.96 62.46 3.93
Sn-Me’ (mm) 0.85 0.97 (0.92, 0.99) 65.48 5.88 60.17 5.43 69.31 6.60 68.62 6.21
Gb.Sn.Pog’ (°) 0.28 0.98 (0.95, 0.99) 13.98 5.67 11.93 5.35 15.08 4.03 15.08 4.43
Cm.Sn.Ls (°) 0.80 0.99 (0.98, 0.99) 92.54 11.44 100.42 10.94 96.88 11.93 97.12 9.04
TFH/FW 0.64 0.75 (0.46, 0.89) 1.01 0.07 0.91 0.05 0.96 0.04 0.94 0.06
IFH/TFH 0.74 0.91 (0.79, 0.97) 0.51 0.02 0.51 0.02 0.51 0.02 0.52 0.02

SD , Standard deviation; CI , confidence interval.

The biometric evaluation was performed on dental casts using a 150-mm digital caliper (CE 03.040487ECC; TÜV Rheinland, Cologne, Germany) with 0.01-mm resolution. Assessments included total mesiodistal diameters of permanent maxillary and mandibular teeth (Σ5-5), intermolar and intercanine widths, diagonal arch length, Little Irregularity index, overjet, overbite, and right canine and left canine relationships ( Fig 3 ). All measurements were obtained by a single examiner, who had been previously calibrated by the ICC.

Fig 3
Mandibular dental arch dimensions measurements obtained from dental casts. 1 , Little Irregularity index; 2 , diagonal arch length; 3 , intercanine width; 4 , intermolar width.

Statistical analysis

After 30 days, the measurements obtained in 20 indigenous individuals selected by stratified random sampling were repeated to verify the error of the method for facial and dental arch biometry. Dahlberg formula was used to determine the random error, and the ICC was used to evaluate the systematic error of measurements.

The ability of the study to include variables identified from each village was examined using the discriminant analysis of the simultaneous method type, separately for males and females, considering the existence of sexual dimorphism of these previously reported variables. ,

Statistical analysis was performed using the BioEstat software (version 5.3; Mamirauá Institute, Belém, Pará, Brazil) for ICC and SPSS statistical software (version 24; IBM, Armonk, NY) for discriminant analysis at 5% level of significance.

Results

A total of 665 indigenous individuals were clinically evaluated. Of this, 567 (85.26%) were excluded from the study because they did not meet the eligibility criteria. Thus, a total of 98 indigenous individuals from 4 remote indigenous villages located in the Medium Valley of the Xingu River, including Arara-Iriri (n = 20; 10 males and 10 females; mean age, 18.9 years), Arara-Laranjal (n = 33; 14 males and 19 females; mean age, 17.8 years), Assurini (n = 25; 12 males and 13 females; mean age, 16.0 years), and Xicrin-Kayapó (n = 20; 11 males and 9 females; mean age, 14.6 years), were recruited to participate in the study.

The initial characteristics of the indigenous population belonging to each of the 4 villages showed an age range that did not interfere in the results of the discrimination. There was a similar distribution between males and females, characterizing the homogeneous groups.

The random and systematic error analysis of the measurements revealed a relatively small error level for the data collected. In the facial analysis, the causal error ranged from 0.13 to 1.81 mm. This difference was always <2% of the total measure. The systematic error revealed excellent replicability, with values ranging from 0.75 to 0.99 ( P <0.0001; Table II ). The random error for the measurements performed on the models was <1 mm for all variables (0.10-0.68 mm). The systematic error as well as the face measurements, demonstrated excellent replicability, with values ranging from 0.94 to 0.99 ( P <0.0001; Table III ).

Table III
Mean, SD, random error and systematic error for intermolar and intercanine widths, diagonal arch length, tooth sizes, Little Irregularity index, overjet, overbite, and right canine and left canine relationships in each indigenous group
Variables Random error (Dahlberg) Systematic error ICC, (95% CI) Arara-Iriri (n = 20) Arara-Laranjal (n = 33) Assurini (n = 25) Xicrin-Kayapó (n = 20)
Mean SD Mean SD Mean SD Mean SD
Maxilla
Intermolar 0.15 0.99 (0.98, 1.00) 54.19 2.73 53.96 2.17 56.45 2.63 53.39 2.46
Intercanine 0.22 0.99 (0.96, 0.99) 37.35 1.82 35.80 1.70 38.51 1.85 36.01 1.41
Arch length 0.59 0.97 (0.92, 0.99) 67.53 4.11 65.01 2.47 68.85 4.30 65.66 4.02
Tooth size (Ʃ15-25) 0.68 0.94 (0.85, 0.98) 75.31 2.90 74.43 3.24 76.53 5.92 74.47 3.15
Little index 0.26 0.98 (0.95, 0.99) 2.23 1.59 2.96 2.26 3.71 2.59 5.12 4.46
Mandible
Intermolar 0.25 0.98 (0.95, 0.99) 46.10 2.04 46.39 2.07 48.21 2.27 45.33 2.25
Intercanine 0.22 0.99 (0.96, 0.99) 27.99 1.40 27.40 1.63 28.65 2.17 27.25 2.22
Arch length 0.40 0.98 (0.96, 0.99) 58.12 2.77 56.43 2.78 58.59 2.83 55.65 3.66
Tooth size (Ʃ35-45) 0.59 0.97 (0.92, 0.99) 64.54 2.63 65.21 2.84 67.71 3.83 64.69 3.32
Little index 0.25 0.99 (0.96, 0.99) 1.85 1.86 2.83 1.79 3.76 2.47 3.43 2.62
General
Overjet 0.14 0.99 (0.99, 0.99) 0.94 1.69 2.27 1.38 1.96 2.52 2.79 1.98
Overbite 0.15 0.99 (0.99, 0.99) 0.52 1.50 1.55 1.12 2.21 1.11 2.66 1.06
Right canine 0.17 0.99 (0.96, 0.99) –0.62 2.30 0.67 1.39 0.45 2.32 1.66 2.60
Left canine 0.10 0.99 (0.97, 1.00) –0.34 1.51 0.43 1.06 0.44 1.90 1.74 2.32
Only gold members can continue reading. Log In or Register to continue

Stay updated, free dental videos. Join our Telegram channel

May 12, 2020 | Posted by in Orthodontics | Comments Off on Dentofacial biometry as a discriminant factor in the identification of remote Amazon indigenous populations

VIDEdental - Online dental courses

Get VIDEdental app for watching clinical videos