Empirical quantification of internal and external rotation muscular co-activation ratios in healthy shoulders

Biomechanical models used to estimate joint loads often predict that antagonistic muscles are inactive or underestimate their contributions [3, 5]. This can result in systematic underestimation of muscle force predictions and joint contact forces. To test the feasibility of employing an empirical co-activation ratio to improve shoulder muscle force modeling estimates, it was purposed to define the co-activation relationship between humeral internal and external rotator muscles in young healthy adults. Electromyography was recorded from rotator cuff and shoulder musculature of 20 adults. Participants performed 54 submaximal voluntary force exertions of humeral internal and external rotation at various humeral abduction and rotation postures. Empirical co-activation relationships for aggregates of humeral internal and external rotators (non-weighted and PCSA-weighted versions) were well characterized by regression models (r (2) = 0.62-0.70) during internal rotation exertions, but only moderately well (r (2) = 0.35-0.42) during external rotation exertions. Humeral abduction and intensity were important predictors in both exertion types. There was no or minimal improvement in r (2) using PCSA-weighted CIs, suggesting low utility. Quantification and implementation of shoulder co-activation into biomechanical models may improve muscle force and joint load estimates, which could assist in more reliable injury risk and tissue load predictions.