Maintenance of muscle strength retains a normal metabolic cost in simulated walking after transtibial limb loss

Abstract

Recent studies on relatively young and fit individuals with limb loss suggest that maintaining muscle strength after limb loss may mitigate the high metabolic cost of walking typically seen in the larger general limb loss population. However, these data are cross-sectional and the muscle strength prior to limb loss is unknown, and it is therefore difficult to draw causal inferences on changes in strength and gait energetics. Here we used musculoskeletal modeling and optimal control simulations to perform a longitudinal study (25 virtual “subjects”) of the metabolic cost of walking pre- and post-limb loss (unilateral transtibial). Simulations of walking were first performed pre-limb loss on a model with two intact biological legs, then post-limb loss on a model with a unilateral transtibial prosthesis, with a cost function that minimized the weighted sum of gait deviations plus metabolic cost. Metabolic costs were compared pre- vs. post-limb loss, with systematic modifications to the muscle strength and prosthesis type (passive, powered) in the post-limb loss model. The metabolic cost prior to limb loss was 3.44±0.13 J/m/kg. After limb loss, with a passive prosthesis the metabolic cost did not increase above the pre-limb loss cost if pre-limb loss muscle strength was maintained (mean -0.6%, p = 0.17, d = 0.17). With 10% strength loss the metabolic cost with the passive prosthesis increased (mean +5.9%, p < 0.001, d = 1.61). With a powered prosthesis, the metabolic cost was at or below the pre-limb loss cost for all subjects with strength losses of 10% and 20%, but increased for all subjects with strength loss of 30% (mean +5.9%, p < 0.001, d = 1.59). The results suggest that maintaining muscle strength may prevent an increase in the metabolic cost of walking following unilateral transtibial limb loss, and that a gait with minimal deviations can be achieved when muscle strength is sufficiently high, even when using a passive prosthesis.

Musculoskeletal Model Personalization Affects Metabolic Cost Estimates for Walking

“…As reported in (Russell, Esposito and Miller, 2018), decreasing the muscle specific tension increased metabolic cost and vice versa. Bhargava's model needed muscle specific tension to be around 37 N/cm 2 rather than 61 N/cm 2 to match experimental averages the best.…”

Rapid predictive simulations with complex musculoskeletal models suggest that diverse healthy and pathological human gaits can emerge from similar control strategies

“…Third, we explored the influence of a transtibial passive prosthesis during walking. To model the prosthesis, we removed the ankle and subtalar muscles (including the gastrocnemii) of the right leg and modelled a passive prosthesis by describing ankle and subtalar torques as linear functions of joint angles q :T=−kq,where k = 800 N m rad −1 is torsional stiffness [36]. We reduced the mass of the lower leg and foot segments by 35% and the moment of inertia by 60% compared to the biological leg [36].…”

Musculoskeletal Model Personalization Affects Metabolic Cost Estimates for Walking

“…To improve CoT predictions further, researchers should consider optimizing the various parameters within metabolic cost models, such as maximum muscle stress. As reported in Russell Esposito and Miller (2018), decreasing the maximum muscle stress value increases metabolic cost and vice versa, which could help reduce magnitude discrepancies in CoT predictions. Interestingly, when (Falisse et al, 2019) used the Bhargava et al (2004) metabolic cost model with different values of maximum muscle stress for different muscles, they obtained reasonable agreement between model predicted and experimentally measured CoT values.…”