Aspects of facial features in a South African sample as applicable for facial approximations

Abstract

Identification of unknown individuals remains a challenge in the South African context. When it is not possible to identify the unknown individual with primary identifiers , e.g. DNA comparisons, forensic facial approximations (FA) are often used. The aim of this study was to evaluate existing guidelines for FA on a sample of black South Africans (BSA). Facial features considered were the eyes, nose, mouth and ears. The study sample consisted of 49 cadavers (38 males and 11 females, mean age 47), 30 computer tomography (CT) scans (23 males, 7 females, mean age 42) and 30 cone beam computer tomography (CBCT) scans (17 males and 13 females, mean age 33) of BSA. Orbital measurements on dissections included the position of the canthi, the diameters and position of the eyeball and the derived height and width of the orbit. Multiplanar level ocular and orbital depth measurements were performed on CT and CBCT scans using MevisLab (www.mevislab.de). The relationship between the features of the anterior nasal spine (ANS) and the shape of the nasal tip was evaluated on dissections. Gerasimov’s two-tangent method for determining the position of the nasal tip was performed on lateral view photographs of the dissected nose. Dental predictors of the width of the mouth and the philtrum were evaluated. Previously published formulae for the prediction of ear length were tested and population specific formulae designed, taking sex and age into consideration. In general, repeatability tests were acceptable apart from those dimensions involving the orbital canal, medial orbital margin and the philtrum. The eyeball was found to be slightly larger than expected, oval-shaped and supero-laterally placed in a rectangular shaped orbit with an eye fissure sloping slightly downwards laterally. A non-projecting nose with a bulbous nasal tip was the most commonly observed, however, no clear association with the features of the ANS was found. The two-tangent method could also not consistently predict the nasal tip position. The cheilia most often corresponded to the canine/firstpremolar junction, while the inter-canine width constituted 60% of the mouth width. No statistically significant correlation could be observed between the upper central incisor width and the flat/absent philtrum width. Therefore, facial approximations guidelines for BSA should include a slightly larger and oval shaped eyeball with an eye fissure sloping slightly downwards laterally and a non-projecting nose with a bulbous tip. A 60% instead of a 75% rule should be used when predicting the mouth width and the cheilia should then correspond to the canine/first-premolar junction. A flat philtrum should be expected. A shorter ear should be used, and the possible influence of sex and age should be taken into consideration when using population specific formulae. With this study possible limitations of the existing guidelines for approximating facial features in BSA were investigated and adaptations suggested. By increasing the sample size, future studies might address the possible influence of sex and aging. Certain landmarks could be revised to improve repeatability and other methodologies should be explored to approximate the nose in BSA.