Aims: To carry out a systematic review of randomized controlled trials (RCTs) to investigate in patients with arthralgia of the temporomandibular joint (TMJ) the effectiveness of TMJ lavage compared to nonsurgical treatment with regard to pain intensity and mandibular range of motion. Methods: The electronic databases Cochrane Controlled Trials Register (1960-2012), PubMed/Medline (1966-2012), and Embase (1966-2012) were systematically searched for relevant RCTs. References of relevant articles were searched for additional studies, as well as citing reports. Two authors independently performed data extraction by using predefined quality indicators. Relevant outcome data included reduction in pain, as assessed by a visual analog scale (VAS) or a pain score, and maximal mouth opening (MMO) before and 6 months after treatment. Included trials were combined using fixed and random effects meta-analysis. Results: Three RCTs (222 patients) were included for meta-analysis. The statistically significant overall standardized mean difference (SMD) (P .001) with regard to pain intensity was -1.07 (95% CI = -1.38, - 0.76) in favor of TMJ lavage. The MMO did not change significantly (P > .05, SMD = .05 [95% CI = -0.33, 0.23]). Conclusions: The results suggest that lavage of the TMJ may be slightly more effective than nonsurgical treatment for pain reduction. However, this difference is not likely to be clinically relevant. Parole chiave: arthralgia, arthrocentesis, arthroscopy, meta-analysis, temporomandibular joint