The aim of this study was to evaluate the relationship between temporomandibular joint (TMJ) effusions and clinical, magnetic resonance imaging and arthroscopic variables. We designed a retrospective cohort study of patients attending our Department who presented temporomandibular joint disease according to the Research Diagnostic Criteria for Temporomandibular Disorders. According to the presence of effusion in magnetic resonance imaging, the sample was divided into 2 groups: patients with or without joint effusion. A total of 203 patients fulfilled the inclusion criteria, of which 99 (48.8%) showed no joint effusion and 104 (51.2%) presented effusion. We found no significant differences between clinical variables and joint effusion. However, patients without effusion showed significantly longer duration of symptoms that patients with effusion (mean = 34.8 months, standard deviation = 43.41 vs 20.25, 23.76; p = 0.005). We found no significant differences in magnetic resonance imaging and arthroscopic variables associated with the presence of effusions. For both groups, there were statistically significant differences in values of mean pain scores and mean maximum interincisal distance between the different times of examination (before surgery and 3, 6 and 12 months after surgery). Joint effusion is not associated with clinical, imaging or arthroscopic variables, but may be a clinical characteristic of early stages of temporomandibular joint disease. Copyright © 2020 European Association for Cranio-Maxillo-Facial Surgery. Published by Elsevier Ltd. All rights reserved.