On March 10, 2019, a systematic search of relevant articles published over the last 20 years was performed in PubMed, as well as in Scopus, the authors’ personal libraries, and the reference lists of included articles. The focus question was: In patients with TMD (P), does TMJ arthrocentesis (I), compared to any control treatment (C), provide positive outcomes (O)?
Thirty papers were included comparing TMJ arthrocentesis to other treatment protocols in patients with disc displacement without reduction and/or closed lock (n = 11), TMJ arthralgia and/or unspecific internal derangements (n = 8), or TMJ osteoarthritis (n = 11). In general, the consistency of the findings was poor because of the heterogenous study designs, and so caution is required when interpreting the meta-analyses. In summary, it can be suggested that TMJ arthrocentesis improves jaw function and reduces pain levels, and the execution of multiple sessions (three to five) is superior to a single session (effect size = 1.82). Comparison studies offer inconsistent findings, with the possible exception of the finding that splints are superior in managing TMJ pain (effect size = 1.36) compared to arthrocentesis, although this conclusion is drawn from very heterogenous studies (I = 94%). The additional use of cortisone is not effective for improving outcomes, while hyaluronic acid or platelet-rich plasma positioning may have additional value according to some studies. The type of intervention, the baseline presence of MRI effusion, and the specific Axis I diagnosis do not seem to be important predictors of effectiveness, suggesting that, as in many pain medicine fields, efforts to identify predictors of treatment outcome should focus more on the patient (eg, age, psychosocial impairment) than the disease.