Background Manual treatments targeting different regions (shoulder, cervical spine, thoracic spine, ribs) have been studied to deal with patients complaining of shoulder pain. Thoracic manual treatments seem able to produce beneficial effects on this group of patients. However, it is not clear whether the patient improvement is a consequence of thoracic manual therapy or a placebo effect. Objectives To compare the efficacy of thoracic manual therapy and placebo thoracic manual treatment for patients with shoulder dysfunction. Methods Electronic databases (MEDLINE, CENTRAL, PEDro, CINAHL, WoS, EMBASE, ERIC) were searched through November 2016. Randomized Controlled Trials assessing pain, mobility and function were selected. The Cochrane bias estimation tool was applied. Outcome results were either extracted or computed from raw data. Meta-analysis was performed for outcomes with low heterogeneity. Results Four studies were included in the review. The methodology of the included studies was generally good except for one study that was rated as high risk of bias. Meta-analysis showed no significant effect for “pain at present” (SMD -0.02; 95% CI: −0.35, 0.32) and “pain during movement” (SMD -0.12; 95% CI: −0.45, 0.21). Conclusion There is very low to low quality of evidence that a single session of thoracic manual therapy is not more effective than a single session of placebo thoracic manual therapy in patients with shoulder dysfunction at immediate post-treatment.