Background Conflicting results exist between somatosensory profiles of patients with temporomandibular myalgia (TMDm). The objective of this review was to examine whether adults with TMDm show altered responses to dynamic quantitative sensory tests compared with healthy controls. Methods We searched five electronic databases for studies, excluding those without suitable controls or where TMDm was associated with confounding non‐musculoskeletal disorders. Risk of bias was assessed with the SIGN case‐control study checklist. Findings were structured around dynamic quantitative sensory tests and their localization. Where possible, we performed meta‐analysis with a random inverse variance model to compare patients with TMDm and healthy controls. Statistical heterogeneity was estimated with Chi² test and inconsistency index, I². Results We extracted data from 23 studies comprising 1284 adults with chronic TMDm and 2791 healthy controls. Risk of bias was assessed as high for 20 studies. Mechanical temporal summation, the most studied phenomenon (14 studies), is increased in the upper limb of patients with TMDm (SMD = 0.43; 95% CI: .11 to .75; p = .009) but not in the jaw area (p = .09) or in the cervical area (p = .29). Very little evidence for altered thermal temporal summation (five studies), conditioned pain modulation (seven studies), exercise‐induced hypoalgesia (two studies), placebo analgesia (two studies), stress‐induced hypoalgesia (one study) and offset analgesia (one study) was found. Discussion A major limitation of this review was the risk of bias of included studies. Future studies would benefit from following methodological guidelines and consideration of confounding factors.