Diagnostic accuracy of paranasal sinus measurements on multislice computed tomography for sex estimation: A systematic review, meta-analysis, and meta-regression

Abstract

This systematic review aimed to answer the question, "Are paranasal measurements obtained with multislice computed tomography able to accurately estimate sex?" This PROSPERO-registered systematic review included six databases and gray literature. Meta-analysis of linear and volumetric data was performed using a random-effects model, mean differences (MDs), and inverse variance. The QUADAS-2 tool-related risk of bias (RoB) and certainty evidence (GRADE approach) were assessed. A total of 1651 articles were initially selected from the main electronic databases for title and abstract screening. After removing duplicates and studies that did not meet the inclusion criteria, 18 studies were eligible for the meta-analysis. The screened studies included 2507 computed tomography of non-pathologic paranasal sinuses of adults aged between 18 and 40 years. Frontal sinus (FS) width (-4.89 mm; p < 0.00001) and maxillary sinus (MS) height (-3.73 mm; p < 0.00001) showed the greatest MD between females and males. Regarding volume, the MD was significant (p < 0.00001) in both paranasal sinuses (-3.01 and -1.04 cm³ , respectively). The best sensitivity (ability to identify males) and specificity (ability to identify females) were found for the anteroposterior (72.3%) and intermaxillary (69%) MS distances, respectively. In general, all studies exhibited low RoB and applicability concerns. The certainty of the evidence was categorized as very low to moderate. Both paranasal sinuses were shown to be dimorphic structures, which could be valuable tools for forensic identification using the FS width and MS anteroposterior distance measurements. However, heterogeneity among studies must be considered.

Keywords: computed tomography; forensic anthropology; frontal sinus; human identification; maxillary sinus; sexual dimorphism.