RESUMO
BACKGROUND: Landing-based measures of the knee are often used to assess risk of anterior cruciate ligament (ACL) injury and inform prevention strategies. There is less understanding of the ankle's influence on knee measures during landing. OBJECTIVE: 1. Examine interactions of dynamic ankle measures alongside various subject and task characteristics on knee dynamics in vertical landing and 2. Determine whether ankle measures alone can estimate dynamic knee measures associated with ACL injury risk. DESIGN: Systematic review and meta-analysis. METHODS: Electronic databases Medline, EMBASE, CINAHL, Web of Science and Cochrane were screened for studies that included measurement of initial contact angles and internal joint moments of both the ankle and knee during landing in uninjured individuals. RESULTS: 28 studies were included for analysis. Using 1144 landing trials from 859 individuals, RRelief F algorithm ranked dynamic ankle measures more important than landing task and subject characteristics in estimating knee dynamics. An adaptive boosting model using four dynamic ankle measures accurately estimated knee extension (R2 = 0.738, RMSE = 3.65) and knee abduction (R2 = 0.999, RMSE = 0.06) at initial contact and peak knee extension moment (R2 = 0.988, RMSE = 0.13) and peak knee adduction moment (R2 = 1, RMSE = 0.00). CONCLUSIONS: Dynamic ankle measures can accurately estimate initial contact angles and peak moments of the knee in vertical landing, regardless of landing task or individual subject characteristics. This study provides a theoretical basis for the importance of the ankle in ACL injury prevention.