BackgroundMultiple investigations have compared the electromyographic (EMG) activity of the scapular muscles between stable and unstable support surfaces during the execution of closed kinetic chain exercises. However, these comparative analyses have grouped different unstable surfaces (wobble board, BOSU, therapeutic ball, and suspension equipment) into a single data pool, without considering the possible differences in neuromuscular demand induced by each unstable support surface. This study aimed to analyze the individual effect of different unstable support surfaces compared to a stable support surface on scapular muscles EMG activity during the execution of closed kinetic chain exercises.MethodologyA literature search was conducted of the Pubmed Central, ScienceDirect and SPORTDiscus databases. Studies which investigated scapular muscles EMG during push-ups and compared at least two support surfaces were included. The risk of bias of included articles was assessed using a standardized quality assessment form for descriptive, observational and EMG studies, and the certainty of the evidence was measured with the Grading of Recommendations Assessment, Development and Evaluation (GRADE) approach. A random-effects model was used to calculate effect sizes (ES, Hedge’s g).ResultsThirty studies were selected in the systematic review. Of these, twenty-three low-to-high quality studies (498 participants) were included in the meta-analysis. The main analyzes revealed, in decreasing order, greater UT EMG activity during push-ups performed on suspension equipment (ES = 2.92; p = 0.004), therapeutic ball (ES = 1.03; p < 0.001) and wobble board (ES = 0.33; p = 0.003); without effect on the BOSU ball. In addition, no effect was observed for SA on any unstable device. The certainty of the evidence ranged from low to very low due to the inclusion of descriptive studies, as well as high imprecision, inconsistency, and risk of publication bias.ConclusionThese findings could be applied in scapular muscles strengthening in healthy individuals. The use of suspension equipment achieves higher UT activation levels. Conversely, the use of any type of unstable devices to increase the activation levels of the SA in shoulder musculoskeletal dysfunctions is not recommended. These conclusions should be interpreted with caution as the available evidence showed a low to very low certainty of evidence, downgraded mostly by inconsistency and imprecision.