A cross-sectional study of the anthropometry of the face among Bonos and Ewes in the Bono region of Ghana

Abstract

Background

Growth and development of craniofacial structures are of importance since their anthropometry is useful in maxillofacial surgery, plastic surgery, orthodontics and forensic medicine. Surgeons consider the specific facial structures of such patients to help obtain accurate results. Ghana is however less-endowed with data regarding facial anthropometry for the aforementioned applications. Therefore, the study aimed to bring out the differences between the facial measurements of the male and female participants in order to generate baseline data for Ghanaians, taking into consideration the Bonos and Ewes living in the Bono region of Ghana. Twenty-four anthropometric facial measurements were taken from a total of 291 healthy individuals (152 Bonos and 139 Ewes), aged 18–60 years using Shahe Vernier callipers.

Results

The study revealed Bono males had significantly longer faces than that of the Bono females (upper face height 2 and total face height). For maxillary height, mandibular width and mouth width, the mean values of Bono males were significantly greater than that of the Ewe males. There was a statistically significant difference concerning nasal length and anatomical nose width among the participants (p < 0.05). Bono male participants had significantly greater values than Ewe males in both endocanthion-exocanthion and endocanthion-endocanthion measurements.

Conclusions

The results of this study support the assertion regarding the existence of tribal variations and sexual dimorphism associated with facial measurements and have provided additional data for facial morphology for biometric and forensic applications as well as facial reconstruction especially among the study populations in Ghana.

Background

The human face is a very useful tool for expression, appearance and identity among others. The features of the face including its bones, muscles, cutaneous and subcutaneous layers all contribute to the unique morphology of a single person (Kumar et al. 2020). The shape of the face is affected by sex, race, ethnicity, climate, socio-economic, nutritional and genetic factors (Jeremić et al. 2013). Anthropometric studies in various populations have detailed the relationships between the landmarks of the face (Arslan et al. 2008; Choe et al. 2004). Males have greater craniofacial measurements such as mouth width, head circumference and minimum frontal breadth than females. Females, on the other hand, have greater cranial base width than males (Anibor et al. 2011). It has been reported that craniofacial measurements can be used in determining stature (Anibor et al. 2011). It is also hypothesized that differences in nose shape may be due to local adaptation to climate (Zaidi et al. 2017). Shetti et al. (2011) found that facial measurements of females show lower mean values compared to males. Males mostly have broader zygomatic regions, supraorbital ridges and prominent mandibles while females may have longer and narrower faces, rounder and broader foreheads and thicker lips (Matthews et al. 2018; Roosenboom et al. 2018).

Pubertal sex hormones also contribute to masculinisation and feminization; the influence of pubertal testosterone causes an increase in lower facial height but a decrease in cheekbone (Hodges-Simeon et al. 2016). It brings about a direct increase in the size and mass of muscles and bones and causes changes in the shape of the face between sexes (Osunwoke et al. 2011). Men usually have less fat tissue; they have stronger and wider bones and more muscle mass compared to women (Marko et al. 2018).

The characteristics of the face can be used as a good identification tool for the dead, missing persons and criminals, using both morphological features and measurements (Ranjana et al. 2016). Growth and development of craniofacial structures are of importance since clinicians depend on them for the process of treatment and diagnosis and in planning for maxillofacial surgeries (Arslan et al. 2008; Tania et al. 2020). Therefore, the knowledge acquired from the study of facial morphology is useful in several areas such as facial aesthetics, forensic medicine and medico-legal disputes as well as reconstructive surgeries (Akinlolu 2016). Surgeons must consider the specific facial structures of such patients to help obtain accurate results (Arslan et al. 2008; Tania et al. 2020). In Ghana, the face is useful in making identity cards and passports among others. Despite the introduction of a biometric system in various identification cards, there is still very little information available on metric facial data and distribution patterns among the tribes. Therefore, the study was designed to bring out the differences between the facial measurements of male and female participants in order to generate baseline data for Ghanaians, taking into consideration the Bonos and Ewes living in the Bono Region of Ghana.

Methods

Study population and location

There are six major ethnic groups in Ghana: the Akans, Ewes, Ga-Adangbes, Mole-Dagbani’s, Guans and Gurmas. Akan is the largest ethnic group in Ghana with several subdivisions. Bono is one of the largest tribes in Ghana and they are predominant in the Bono region of Ghana. The largest tribe in the Volta region of Ghana is the Ewes. A significant number of the Ewes are also found in the Bono region. The study took interest in both tribes since they have completely different ancestral origins and physical features that can help make our comparisons between both tribes in the Bono region of Ghana.

This quantitative cross-sectional study employed a non-probability purposive sampling technique to recruit a total of 291 healthy individuals (162 males and 129 females). The study and its protocols were explained to the understanding of the participants. The study adhered to the Helsinki Declaration and was approved by the Committee on Human Research, Publication and Ethics of the University. Informed participant consent was obtained from each participant before facial measurements were taken.

The sample size was determined using the Yamane formula (Adam 2020).

$$n=\frac{N}{1+N{\left(\propto \right)}^2}$$

where n = sample size

N = study population = 1525

α = margin of error = 0.05; confidence level of 95%

Therefore,

$$n=\frac{N}{1+N{\left(\propto \right)}^2}$$

$$\frac{1525}{1+1525{(0.05)}^2}$$

$$n=316$$

However, the sample size of 316 could not be achieved due to factors such as outliers and unwillingness of some individuals to take part in the study.

Inclusion and exclusion criteria

Healthy participants with both parents and grandparents being Bono or Ewe and living in the Bono region were included in the study. Individuals with physical impairment, scarring, craniofacial trauma, amputated limbs, visible tumours, facial oedema and pregnant women were excluded from the study. Participants with physical signs of endocrine disorders such as dwarfism and gigantism were also excluded from the study.

Data collection

The Shahe Vernier calliper was used to take the 24 anthropometric facial measurements (to the nearest 0.1 mm). Manual anthropometry using the digital Shahe Vernier calliper is a gold standard since it is easily accessible. It can access the various bony landmarks directly and it is less costly.

The 24 measurements were taken since they are easily identifiable, palpable features on the face of an individual (Figs. 1 and 2). All measurements were duplicated and the averages were taken. To avoid inter-observer error, all the measurements were taken by the same person. All anthropometric measurements of the face were taken with the participants in sitting position with the back perpendicular to the chair, body erect, head up and arms at the sides. A pilot study done prior to the main study showed that eye fissure width, ear width and ear length were bilaterally symmetrical so measurements were taken from one side specifically the left part of the face. The various facial measurements taken from the participants have been explained in Table 1.

Fig. 1

Measurements of the lateral aspect of the face: upper face height two (tr-n); total face height (tr-gn); morphological face height (n-gn); nasal height (n-sn); ear length: sa-sba (Farkas et al. 2005)

Fig. 2

Measurements on the frontal aspect of the face: intercanthal width (en-en); biocular width (ex-ex); eye fissure width (en-ex); face breadth (zy-zy); mandibular width (go-go); morphological nose width (al-al); mouth width (ch-ch) (Farkas et al. 2005)

Table 1 Facial measurements taken from the participants and their meanings

Data analysis

Statistical analyses were carried out using International Business machines (IBM), Statistical Package for Social Sciences (SPSS) (version 24.00, incorporated, (Inc.) Chicago, Illinois (IL), United States of America (USA)). The measurements were expressed as means and standard deviations. Differences between tribes were tested using ANOVA, followed by Tukey’s Multiple Comparison (TMC) test. The results of the post hoc tests were indicated in the tables by letter indices ‘a, b, c, d, e, and f’, for pairs of groups with a significant difference. The p values are only the results of the ANOVA (overall difference among the four groups) and not the post hoc test.

Results

Analysis of variance of facial anthropometric data stratified by sex within the studied tribes

From the post hoc analysis using Tukey’s honest significance difference (HSD), there were no significant differences for upper face height 1 (tr-g) between Bono males, Bono females, Ewe males and Ewe females. Post hoc analysis revealed a significant difference between Bono males and Bono females, with the Bono males having higher values than Bono females in both upper face height 2 (tr-n) (p = 0.014) and total face height (tr-gn) (p = 0.032) (Table 2).

Table 2 Analysis of variance of the upper facial measurements of the Bono and the Ewe participants

All morphological facial measurements showed significant variations between sexes among the tribes. In both mid-facial height 1(n–st) and mid-facial height 2 (g-sn), significant differences were observed between Bono males and Bono females (p = 0.023) following post hoc test. For morphological face height, maxillary height, face breadth, mandibular width and mouth width, the values of Bono males were significantly greater than those of the Ewe males following post hoc analysis. In the case of mandibular height, post hoc analysis showed that a significant difference was between Bono females and Ewe males (Table 3).

Table 3 Analysis of variance of morphological facial measurements of Bono and Ewe participants

For the nasal measurements, only anatomical nose width(ac-ac) and nasal length (n-pn) showed some significant differences among the studied populations. Anatomical nose width was significantly different between Bono females and Bono males as well as Bono males and Ewe females. However, for nasal length, it showed a significant difference only between Bono males and Ewe males (Table 4).

Table 4 Analysis of variance (ANOVA) of nasal measurements of Bono and Ewe participants

The means of the intertragal width (t-t) for the Bono males and Ewe males were significantly higher than those for the Bono females and Ewe female participants. One-way analysis of variance of intertragal width, tragus-exocanthion, tragus-subnasale, tragus-cheilion and tragus-gnathion among Bono males, Bono females, Ewe males and Ewe females showed statistically significant differences. The post hoc analysis revealed that the significant difference was between Bono females and Bono males, Bono males and Ewe females for the tragus to tragus (t-t) (p = 0.007) (Table 5).

Table 5 Analysis of variance of tragal-related measurements by tribes

The mean tragus-exocanthion (t-ex) of the Bono male and female participants was statistically different from Bono female and Ewe male, Bono male and Ewe female, Ewe female and Ewe male participants (p = 0.004). The post hoc analysis revealed a significant difference between Bono female and Ewe male participants for tragus-subnasale (p=0.009). There were significantly higher values for Bono male and Ewe male participants. The post hoc analysis revealed significant differences between Bono females and Ewe males, Ewe females and Ewe males for the mean of tragus-cheilion measurements. The mean of tragus to gnathion was statistically significant for Bono female and Bono male participants (p = 0.017) (Table 5).

From the post hoc analysis using Tukey honest significant difference, there were no significant differences in biocular width (ex-ex) among tribes and sex. However, for eye fissure width and intercanthal width, there was a statistically significant difference between Bono males and Ewe males. The Bono male participants had significantly higher values than Ewe males in both endocanthion-exocanthion and endocanthion-endocanthion (Table 6).

Table 6 Analysis of variance of ocular measurements stratified by tribe

From Table 7, all the auricular measurements showed significant variation between sex within the same tribe. The post hoc analysis revealed that in both ear length (sba-sa) and ear width (pra-pa), significant differences were observed between Bono females and Bono males at p = 0.012 and p = 0.022 respectively.

Table 7 Analysis of variance of auricular measurements

Discussion

Facial anthropometric measurements stratified by sex and tribe

In the present study, although the male participants recorded numerically greater upper facial anthropometric measurements than the females, those that showed significant sexual dimorphism were upper facial height 2 and total facial height. This could be attributed to pubertal testosterone which causes a steep and prolonged increase in lower facial height, size and mass of muscles in males than females resulting in differences in facial measurements (Osunwoke et al. 2011; Marko et al. 2018) as well as bony structures, soft-tissue, subcutaneous fat and fascia of the skin are affected by the ageing process. However, subcutaneous fat muscle and fascia act in dynamic unison to determine the phenotypic presentation of the face throughout life (Zimbler et al. 2001).

The present study demonstrated sexual differences regarding total facial height which agrees with the works of Dayal et al. (2008), Singla et al. (2020) and Maalman et al. (2017). This could be due to the similarity in climate, diet, ethnicity and socio-economic factors. Total facial height is likely to be one of the major variables important for discriminating between males and females (Patil and Mody 2005).

The mandibular width of the present study was significantly different from that of Northwest Indians (Sahni et al. 2010). This difference could be a result of genetic trait of morphological facial measurements. Moreover, the maxillary height of adult Ibibios of Nigeria according to Oladipo et al. (2010) and Didia and Dapper (2005) also in Nigeria exhibited sexual dimorphism which agrees with the present study and could be due to genetic factors and similarity in race since Ghanaians and Nigerians are both Negroes, and this might have brought about the existing differences in shape and configuration of the maxillary height of both sexes.

The present study agrees with the findings of Ewunonu and Anibeze (2013) in the South Eastern Nigerian population where the mean nasal height and width among the male participants were significantly higher than that of the female participants. Shetti et al. (2011) found that nasal measurements of females show lower mean values compared to males. Zaidi et al. (2017) hypothesized that differences in nose shape may be due to local adaptation to climate and reported that the distance between nasal alare was significantly higher among West Africans, South Asians and East Asian ancestry compared to the European ancestry. Facial anthropometric measurements show racial variations according to nasal type. African noses appear to be the shortest and widest. Afro-Caucasians have the narrowest, with Afro-Indians having the longest. Ageing causes significant changes in the appearance of the nose and nasal elongation (Porter and Olson 2001).

The nose is an individual’s most defining feature because it is at the centre of the face, anthropometric analysis of it can provide data which could contribute to satisfactory results of cosmetic nasal surgery. Nasoplastic surgeons require access to facial databases based on accurate anthropometric measurements to perform optimum correction in both sexes. A successful outcome in rhinoplasty requires a thorough and accurate preoperative planning and awareness of the morphological differences (Ofodile and Bokhari 1995).

All the tragal-related measurements of both tribes such as intertragal width, tragus-exocanthion, tragus-subnasale, tragus-cheilion and tragus-gnathion were sexually dimorphic and males recorded greater values than females since they have stronger and wider bones and more muscle mass compared to females (Anibor et al. 2011; Marko et al. 2018). This indicates that facial equipment such as face masks, helmets and spectacles must be designed considering sex. Lee et al. (2012) recommended that Korean female facial characteristics need to be considered in the design of oxygen masks to fit Korean pilots. In the present study, eye measurements, mean intercanthal width and biocular width were numerically greater in males than in females. This is supported by the findings of Matthews et al. (2018), who reported that males mostly have broader supraorbital ridges than females which might have enhanced the intercanthal and biocular width to be greater in males than the females. However, eye fissure width showed no significant difference between males and females. This could be attributed to minimal changes in facial growth at age range 18–60 years in some of the facial measurements (Didia and Dapper 2005).

Moreover, all the auricular measurements, ear width and ear length showed significant differences between Bono males and Bono females. This could be attributed to auricular expansion which starts earlier in males than females (Taura et al. 2013). In contradiction to this finding, Maalman et al. (2017) reported that Dagaaba and Sisaala female participants in Ghana have greater ear length than the males, attributing the differences to the wearing of heavy earrings by the females and a possible loss of fat as one advances in age. The mean ear length and width of the Bono and Ewe males were numerically lower than those of a South-Eastern Nigerian population (Ewunonu and Anibeze 2013). Also, in a study by Japatti et al. (2018) among Maharashtrian adults in India, there was sexual dimorphism in ear width but their mean values were not statistically significant. Several authors have stated that ear length increases faster and for longer duration compared to ear width (Purkait and Singh 2007; Meijerman et al. 2007; Niemitz et al. 2007; Japatti et al. 2018). This could be attributed to the human external ear which grows continuously even after skeletal maturity is reached. Ageing brings about changes in the microscopic structure of the cartilage and this decreases elastic fibres and density of the cartilage causing an increase in length and reduction in the breadth of the ear (Japatti et al. 2018).

Conclusions

Male facial measurements were numerically higher than those of the female participants. There were no significant differences for upper face height one among males and females of the Bono and Ewe tribes. All the measurements classified under morphological facial and auricular showed significant variations among the tribes stratified by sex. There was a statistically significant difference concerning nasal length and anatomical nose width among the participants (p < 0.05). The Bono male participants had significantly higher values than Ewe males in both endocanthion-exocanthion and endocanthion-endocanthion.

The results of this study support the assertion regarding the existence of tribal variations and sexual dimorphism associated with facial measurements and have provided additional data for facial morphology for biometric and forensic applications as well as facial reconstruction especially among the study populations in Ghana.