Abstract
Long bones play an important role as an aid in sexual estimation. The humerus can be an important alternative in human identification, when bones such as skull and pubis are not available. Osteometric methods are inexpensive and accurate, but are population-specific. Therefore, the aim of the present study was to evaluate the accuracy of the humerus estimating sex in a population in southeastern Brazil. The study used a sample composed of 42 cataloged humeral bones, 30 of men and 12 of women, which are part of the Biobank "Bones, teeth and human cadavers" of the Anatomy laboratory of the Piracicaba School of Dentistry (UNICAMP). Seven parameters of each humerus bone were measured using a digital caliper and an osteometric table. Differences in measurements were subjected to standard statistical analysis and a discriminating point was determined to indicate whether the bone is male or female. Sexual dimorphism ratios were calculated to determine the level of differences between the sexes. The difference in means for all variables was significantly greater in males compared to females (P<0.01) with the most effective parameter for predicting gender being the width of the epicondyle with an accuracy of 85.7%. The results indicated good precision in one of the studied parameters, even with the limited sample size, suggesting that the humerus is a bone that can be used in this population. We conclude that the humerus is an effective bone that can be safely used to estimate sex in this Brazilian population, but studies with larger sample sizes need to be performed.