Potential of Individual Upper-Limb Muscles to Contribute to Postural Tremor: Simulations From Neural Drive to Joint Rotation.

Background: It is unclear which muscles contribute most to tremor and should therefore be targeted by tremor suppression methods. Previous studies used mathematical models to investigate how upper-limb biomechanics affect muscles' potential to generate tremor. These investigations yielded principles, but the models included at most only 15 muscles. Here we expand previous models to include 50 upper-limb muscles, simulate tremor propagation, and test the validity of the previously postulated principles. Methods: Tremor propagation was characterized using the gains between tremorogenic neural drive to the 50 muscles (inputs) and tremulous joint rotations in the 7 joint degrees-of-freedom (DOF) from shoulder to wrist (outputs). Each gain can be interpreted as the potential of a muscle to generate tremor in a DOF. Robustness and sensitivity analyses were performed to assess the effects of model parameter variability on gains. Results: Simulations of postural tremor using the expanded model confirmed the previously postulated principles and revealed new insights, including: 1) most of the muscles with the largest gains were among the 15 muscles in the original model; 2) some gains depended strongly on posture; 3) averaged across the postures included in this study, the largest gains belonged to input-output pairs involving biceps/forearm/wrist muscles and forearm/wrist DOF, 4) although some shoulder and extrinsic hand muscles also exhibited large gains, especially in select postures. Discussion: These observations suggest that in the absence of additional information (such as tremorogenic neural drive to muscles), peripheral tremor suppression efforts should start by targeting biceps/forearm/wrist muscles or forearm/wrist DOF.