RESUMO
BACKGROUND: This study aimed to identify the postural sway variables that can distinguish between ballet dancers with high and low occurrences of musculoskeletal injuries. METHODS: Fourteen professional ballet dancers were assigned to a high-occurrence group (N = 5, reports >2 injuries in the previous 6-months) or a low-occurrence group (N = 9, reports ≤1 injury). Center-of-pressure (COP) data were acquired using a force platform during the following tasks: single-leg stance with eyes open, single-leg stance with eyes closed, and demi-pointe stance with eyes open. The COP standard deviation (SD) and range (RA) in both the medial-lateral (ML) and anterior-posterior (AP) directions were estimated. Between-group comparisons were made through Welch's t-tests for unequal sample size along the effect size measure (Cohen's d). Spearman's rho was used to estimate the association between the number of injuries and the COP variables. The statistical threshold was set at 1%. RESULTS: A between-group effect was found only for the demi-pointe stance, with large effects for SDML (P = 0.006, d = 1.7), RAAP (P = 0.006, d = 1.7), and RAML (P = 0.005, d = 1.7). An inverse relationship was found between the number of injuries and the demi-pointe's COP range in both directions (Spearman's rho from -0.681 to -0.726, P = 0.007). CONCLUSIONS: COP measures taken in ballet-specific positions can distinguish between dancers with a high and low occurrence of musculoskeletal injuries. Suggestions are made to include ballet-specific tasks in the functional assessments of professional dancers.