Effects of Variability in Triceps Surae Muscle Volumes on Triceps Surae Muscle Forces During Human Walking

Abstract Body : Musculoskeletal modelling allows for the estimation of muscle forces during locomotion, which can aid in interpreting skeletal morphology. These models, however, depend on underlying parameters, such as muscle volume, pennation angle, and fiber length, to solve the muscle redundancy problem and calculate individual muscle forces. Muscle volume distribution among the triceps surae muscles (gastrocnemius lateralis, gastrocnemius medialis, and soleus) vary across humans. Here we test how variation in the distribution of muscle volumes across triceps surae impacts maximum gastrocnemius lateralis, gastrocnemius medialis, and soleus muscle forces during walking.

Three patterns of triceps surae muscle volume distributions (average of 21 young, healthy adults, average of 21 cadaveric specimens, and the cadaveric specimen that is standard for the AnyBody modeling system) were evaluated in a standard musculoskeletal model using kinetics and kinematic data of two healthy individuals (one male, one female) at three velocities. Maximum muscle forces during propulsion and braking were calculated in the AnyBody modeling system using inverse dynamics. Kruskal-Wallis tests were used to test for significant differences among the three patterns of muscle distribution during walking within each of the two subjects (separately) at each velocity.

Muscle volume distribution significantly affects gastrocnemius lateralis and soleus muscle forces during propulsion in both individuals at all velocities. Gastrocnemius lateralis forces are significantly higher when the young, healthy muscle volume distribution is used (average maximum difference 344.04N; p< 0.007), while soleus forces are significantly higher when the standard cadaveric model distribution is used (average maximum difference 516.49N; p< 0.005). During braking, both soleus and gastrocnemius lateralis significantly differ at all speeds in the female (p< 0.035) and at the controlled normal speed in the male (p=0.007), with gastrocnemius lateralis again being higher when the young distribution is used (average maximum difference 131.49N) and soleus higher when the standard cadaveric distribution is used (average maximum difference 198.60N). Gastrocnemius medialis only differs during the propulsion phase in the male (at all speeds; average maximum difference 267.99N; p< 0.020). These results suggest that researchers should pay careful attention to the muscle parameters used in their models. This is particularly relevant if they are interested in using muscle force data to understand morphological variation because the distribution of strain across skeletal elements is inextricably intertwined with the forces that act on the bone.