Asymmetries in individuals with lower limb amputation are not explained by muscle co-contraction. A pilot study

Introduction: Maintaining upright posture in humans is challenging because of the high centre of mass and the small base of support. Muscle activation is modulated according to task demands and body orientation relative to gravity. Studying the effect of changes in gravity allows the identification of weight-bearing mechanisms that can inform rehabilitation and technical solutions (e.g., prosthesis) for humans on Earth and in space. In the laboratory, changes in body orientation, such as head-up or head-down tilt, can alter the axial body loading. In this study, we explored balance strategies in individuals with lower-limb amputation under different body loading conditions to (i) understand the nature of the compensatory adjustments often observed in standing and locomotion (Murray et al., 2017) and (ii) the potential adaptations in altered gravity. The aim of this preliminary study was to explore how balance and lower-limb muscle activation vary across body orientations in individuals with lower-limb amputation to explore neuromuscular control strategies following limb loss.Methods: This study was part of a larger project and the data from four people with unilateral lower limb amputation (1 female, 3 males, mean ± std age 51 ± 17 y, height 1.72 ± 0.11 m, weight 76 ± 15 kg, time since amputation 13 ± 9 y; cause: congenital, trauma n=2, vascular; side 2 left, 2 right; 1 below knee, 3 above knee) and from age and sex matched controls (1 female, 2 males, 47 ± 22 y,1.80 ± 0.10 m, 72 ± 19 kg; all right dominant) are reported. Where possible, bilateral electromyography (EMG) recordings (1320 Hz) were recorded from the tibialis anterior (TA), gastrocnemius medialis (GM), vastus lateralis (VL), and semitendinosus (ST). Muscle activation was recorded while participants were standing (1 g) on two force platforms (1080 Hz) and then on a motorised plinth in three body orientations to simulate changes in the gravitational loading along the head-foot body axis: supine at 0 deg (0 g), 6 deg head-down tilt (-0.10 g), and 60 deg head-up tilt (0.87 g). Each trial lasted 30 s.Antero-posterior (AP) and medio-lateral (ML) centre of pressure (CoP) sway range were calculated. A one-tailed Wilcoxon Signed-Rank test was used to determine the sway differences between limbs in each group. The Robinson Symmetry Index was calculated for CoP and the vertical component of the force vector:SI = [(side1 – side2) / 0.5*(side1 + side2)]*100where side1 was the dominant side for controls and the unaffected side for our test participants. SI=0 indicates perfect symmetry, SI>0 indicates higher sway or force on the unaffected from dominant side, SI<0 indicates higher sway or force from the affected or non-dominant side. A Wilcoxon Rank-Sum test was applied to test differences in SIs between the two test groups.EMG root mean square was calculated and then normalised to the maximum in each condition. Agonist-antagonist co-contraction indexes (CCI) were calculated as the area under the overlapping curve (TA-GM, VL-ST), divided by the number of samples T (Unnithan and Frost CCI = Overlap_area /T). Kruskal-Wallis analysis was applied to test differences in CCI according to body loading. Pearson’s correlations between CCI and sway and force data were run for the standing trials.Results: Despite the small sample size, we found differences which were close to significance. AP and ML sway were higher in the unaffected limb in test participants, but no difference was found in controls. Only the unaffected/dominant side showed larger ML sway in the test group (unaffected/dominant ML p=0.057; AP p=0.63; affected/non dominant ML p=0.86; AP p=0.23). AP and ML SI suggest higher asymmetries in the test group, with larger sway in the unaffected limb. Vertical force SIs were 22.39 ± 12.69 % for the participants with amputation and 6.03 ± 9.32 % for controls (p=0.23).No difference was found in CCI in the unaffected side for participants with an amputation. No significant correlations were found, despite some high R2 (vertical force SI vs. VL-ST CCI b=0.0011, R2=0.8947; AP sway vs. VL-ST CCI b=-0.0004, R2=0.6805; AP sway vs. TA-GM CCI b=-0.0004, R2=0.6768).Conclusions: This study suggests sway asymmetries in people with lower limb amputations. All the asymmetry indexes indicated higher force and sway in the unaffected/dominant limb. The co-contraction index calculated during standing was related to higher vertical force asymmetry, and lower AP sway asymmetry.The unaffected limb is preferential for sway control. However, muscle co-contraction alone does not explain the prominent role of the unaffected limb. These results reflect adaptative control mechanisms to cope with varying weight-bearing demands, especially in participants with a well-adjusted prosthesis and years of practice, as in our sample. However, varying axial body loading shows adaptations needed under less familiar conditions. These adaptations may be relevant for those who rely on weight-bearing assistive devices, but also for the participation of individuals with amputation in future spaceflight missions. In space, adaptable control strategies are required, and investigating the control mechanisms beyond muscle co-contraction is needed to manage the variations in loading and prosthesis’ fit and use within a mission.Acknowledgements: This study was funded by the UK Space Agency (UKSA) and the Science and Technology Facilities Council (STFC), grant number UKRI3372. We thank all the participants who volunteered for this study.