Osteometric analysis for sexing of modern sternum – An autopsy study from South India

• Sexing of modern autopsied sternum was carried out using statistical considerations.
• Sex differences were observed for all linear measurements of sternum.
• Predictability of sternum in sexing using univariate models ranged between 67.5% and 74.4%.
• Classification accuracy rates were higher in multivariate analysis.
• MB along with S1 yielded maximum sexing accuracy of 79.5% and 81.2% using discriminant function and logistic regression analysis.

Estimation of sex is considered as one of the essential parameters in forensic anthropology and requires foremost attention in the analysis of commingled and unidentified remains. In India, there is a paucity of population specific morphometric standards for identifying sex from unknown human remains in different population groups. The present research is an osteometric analysis to study the sexual dimorphism of the sternum of South Indian origin using statistical considerations. The study sample constituted of adult autopsied sternums of known age and sex. Five linear measurements (length of the manubrium, mesosternum, manubrium and mesosternum together, and width at 1st and at 3rd sternebra) were examined during the study. Three indices (manubrio-corpus Index, ratio of the length of the mesosternum and manubrium, and sternubrial-width index) were computed from the length and width measurements of the sternum. Statistical analysis was done using SPSS computer software and Student’s t-test was applied to find the sex differences in these variables. While statistically significant sex differences were observed for all the five linear measurements of the sternum, none of the sternal indices showed statistically significant sex differences. Discriminant function and logistic regression analysis were performed to derive the predicting models for estimation of sex from the different variables. The predictability of sternal measurements in sexing using univariate models ranged between 67.5% for the width at 3rd sternebra and 74.4% for the combined length of manubrium and mesosternum. The classification accuracy rates of sternal measurements were observed to be higher when multivariate analysis was performed. Length of manubrium and mesosternum together along with width at 1st sternebra yielded maximum accuracy of 79.5% (discriminant function analysis) and 81.2% (logistic regression analysis) in sexing of male and female sternum. The present research concludes that the application of sternum in sex estimation should be restricted to cases when other more reliable bones for sexing are not available to the investigators.