The objective of this study was to systematically evaluate gender differences in the prevalence of TMD. A systematic review was performed in PubMed, EMBASE, Web of Science and LILACS in duplicate by two independent reviewers. The inclusion criteria were cross-sectional studies that reported the prevalence of TMD for men and women and that used the Research Diagnostic Criteria for Temporomandibular Disorders (RDC/TMD) Axis I group diagnostic criteria:(group I = muscle disorders; group II = disc displacements; group III = arthralgias/arthritis/arthrosis).To be eligible for inclusion, studies must include adult individuals (>18 years) from a non-clinical population (ie without pre-diagnosis of TMD); in other words, from population-based studies. There were no restrictions on the year and language of publication. The quality of the articles was assessed by an adapted version of the Newcastle-Ottawa Scale(NOS), and the publication bias was assessed by a funnel plot graph. Data were quantitatively analysed by meta-analysis using odds ratio (OR) as the measure effect. The electronic search retrieved a total of 6104 articles, of which 112 articles were selected for full-text reading according to the eligibility criteria. By means of manual search, one study was retrieved. Five articles were selected for meta-analysis with a combined sample of 2518 subjects. Women had higher prevalence of TMD in all RDC/TMD diagnostic groups. The meta-analysis yielded the following results: (a) OR = 2.24 for global TMD (groups I, II and III combined), (b) OR = 2.09 for group I, (c) OR = 1.6 for group II and (d) OR = 2.08 for group III. The importance of gender in the development of TMD has been demonstrated, with a two times greater risk of women to develop it as compared to men.