Osteometric sex assessment on shoulder region: a radiodiagnostic study

Sex estimation from skeletal remains is a fundamental component of forensic and anthropological investigations. Although the pelvis is considered the most reliable skeletal element for sex estimation due to its pronounced sexual dimorphism, it may be unavailable or unsuitable for analysis in fragmented or incomplete remains. This study aimed to evaluate the osteometric utility of the shoulder complex as an alternative anatomical region for sex estimation.

This retrospective study analyzed computed tomography (CT) images of 217 adults (113 males, 104 females) aged 18-55 years. A total of 12 morphometric measurements and one proportional index were obtained from the humerus, scapula, and clavicle. Multivariable logistic regression and receiver operating characteristic (ROC) analyses were performed to identify significant predictors and determine optimal cut-off values for sex estimation.

The area under the ROC curve for the humeral head diameter and glenoid cavity width variables were found to be statistically significant. When a cut-off value of 41.95 mm was applied to the measurements of the humeral head diameter variable, the observed sensitivity was 83.2% and specificity was 88.5%, while a cut-off value of 27.85 mm was applied to the measurements of the glenoid cavity width, the observed sensitivity was 82.3% and specificity was 90.4%.

Humeral head diameter and glenoid cavity width demonstrated strong discriminatory performance for sex estimation. These CT-based shoulder complex measurements provide reliable, population-specific cut-off values and may serve as practical tools in forensic and anthropological cases involving fragmented skeletal remains or isolated upper extremity structures.