Comparing spinal loads in individuals with unilateral transtibial amputation with and without chronic low back pain: An EMG-informed approach.

Chronic low back pain (cLBP) is highly prevalent after lower limb amputation (LLA), likely due in part to biomechanical factors. Here, three-dimensional full-body kinematics and kinetics during level-ground walking, at a self-selected and three controlled speeds (1.0, 1.3, and 1.6 m/s), were collected from twenty-one persons with unilateral transtibial LLA, with (n = 9) and without cLBP (n = 12). Peak compressive, mediolateral, and anteroposterior L5-S1 spinal loads were estimated from a full-body, transtibial amputation-specific OpenSim model and compared between groups. Predicted lumbar joint torques from muscle activations were compared to inverse dynamics and predicted and measured electromyographic muscle activations were compared for model evaluation and verification. There were no group differences in compressive or anterior shear forces (p > 0.466). During intact stance, peak ipsilateral loads increased with speed to a greater extent in the cLBP group vs. no cLBP group (p=0.023), while during prosthetic stance, peak contralateral loads were larger in the no cLBP group (p=0.047) and increased to a greater extent with walking speed compared to the cLBP group (p=0.008). During intact stance, intact side external obliques had higher activations in the no cLBP group (p=0.039), and internal obliques had higher activations in the cLBP group at faster walking speeds compared to the no cLBP group. Predicted muscle activations demonstrated similar activation patterns to electromyographic-measured activations (r = 0.56-0.96), and error between inverse dynamics and simulated spinal moments was low (0.08 Nm RMS error). Persons with transtibial LLA and cLBP may adopt movement strategies during walking to reduce mediolateral shear forces at the L5-S1 joint, particularly as walking speed increases. However, future work is needed to understand the time course from pain onset to chronification and the cumulative influence of increased spinal loads over time.

Linking whole-body angular momentum and step placement during perturbed human walking.

Human locomotion is remarkably robust to environmental disturbances. Previous studies have thoroughly investigated how perturbations influence body dynamics and what recovery strategies are used to regain balance. Fewer studies have attempted to establish formal links between balance and the recovery strategies that are executed to regain stability. We hypothesized that there would be a strong relationship between the magnitude of imbalance and recovery strategy during perturbed walking. To test this hypothesis, we applied transient ground surface translations that varied in magnitude, direction and onset time while 11 healthy participants walked on a treadmill. We measured stability using integrated whole-body angular momentum (iWBAM) and recovery strategy using step placement. We found the strongest relationships between iWBAM and step placement in the frontal plane for earlier perturbation onset times in the perturbed step (R2=0.52, 0.50) and later perturbation onset times in the recovery step (R2=0.18, 0.25), while correlations were very weak in the sagittal plane (all R2≤0.13). These findings suggest that iWBAM influences step placement, particularly in the frontal plane, and that this influence is sensitive to perturbation onset time. Lastly, this investigation is accompanied by an open-source dataset to facilitate research on balance and recovery strategies in response to multifactorial ground surface perturbations, including 96 perturbation conditions spanning all combinations of three magnitudes, eight directions and four gait cycle onset times.

Which lower limb joints compensate for destabilizing energy during walking in humans?

Current approaches to investigating stabilizing responses during locomotion lack measures that both directly relate to perturbation demands and are shared across different levels of description (i.e. joints and legs). Here, we investigated whether mechanical energy could serve as a 'common currency' during treadmill walking with transient unilateral belt accelerations. We hypothesized that by delivering perturbations in either early or late stance, we could elicit net negative or positive work, respectively, from the perturbed leg at the leg/treadmill interface, which would dictate the net demand at the overall leg level. We further hypothesized that of the lower limb joints, the ankle would best reflect changes in overall leg work. On average across all seven participants and 222 perturbations, we found early stance perturbations elicited no change in net work performed by the perturbed leg on the treadmill, but net positive work by the overall leg, which did not support our hypotheses. Conversely, late stance perturbations partially supported our hypotheses by eliciting positive work at the leg/treadmill interface, but no change in net work by the overall leg. In support of our final hypothesis, changes in perturbed ankle work, in addition to contralateral knee work, best reflected changes in overall leg work.

Onset timing of treadmill belt perturbations influences stability during walking.

Split-belt treadmills have become popular tools for investigating stability during walking by using belt accelerations to induce slip-like perturbations. While the onset timing of destabilizing perturbations is a critical determinant of an individual's stabilizing response, previous studies have predominantly delivered belt acceleration perturbations at heel strike or have not explicitly controlled onset as a percentage of the gait cycle. To address this gap, we 1) developed an algorithm to target transient increases in unilateral belt speed to begin at specific percentages of the walking gait cycle, 2) validated the algorithm's accuracy and precision, and 3) investigated the influence of different onset timings on spatial stability measures. We evaluated desired onset timings of 10, 15, 20, and 30% of the gait cycle during walking at 1.25 m/s and measured step lengths and widths, as well as anteroposterior and mediolateral margins of stability during the perturbed and four recovery steps in 10 able-bodied participants. From 800 perturbations, we found a mean (standard deviation) delay in onset timing of 5.2% (0.9%) of the gait cycle, or 56 (9) ms. We hypothesized later onset timings would elicit more stabilizing responses due to the less stable configuration of the body during late vs. early single stance. Our data generally supported this hypothesis - in comparison to earlier onset timings, later onset timings precipitated greater stabilizing responses, including larger step lengths, step widths, and anteroposterior/mediolateral margins of stability on the perturbed step, in addition to shorter step lengths and wider step widths on the first step post-perturbation.

Adding carbon fiber to shoe soles may not improve running economy: a muscle-level explanation.

In an attempt to improve their distance-running performance, many athletes race with carbon fiber plates embedded in their shoe soles. Accordingly, we sought to establish whether, and if so how, adding carbon fiber plates to shoes soles reduces athlete aerobic energy expenditure during running (improves running economy). We tested 15 athletes as they ran at 3.5 m/s in four footwear conditions that varied in shoe sole bending stiffness, modified by carbon fiber plates. For each condition, we quantified athlete aerobic energy expenditure and performed biomechanical analyses, which included the use of ultrasonography to examine soleus muscle dynamics in vivo. Overall, increased footwear bending stiffness lengthened ground contact time (p = 0.048), but did not affect ankle (p ≥ 0.060), knee (p ≥ 0.128), or hip (p ≥ 0.076) joint angles or moments. Additionally, increased footwear bending stiffness did not affect muscle activity (all seven measured leg muscles (p ≥ 0.146)), soleus active muscle volume (p = 0.538; d = 0.241), or aerobic power (p = 0.458; d = 0.04) during running. Hence, footwear bending stiffness does not appear to alter the volume of aerobic energy consuming muscle in the soleus, or any other leg muscle, during running. Therefore, adding carbon fiber plates to shoe soles slightly alters whole-body and calf muscle biomechanics but may not improve running economy.

A more compliant prosthetic foot better accommodates added load while walking among Servicemembers with transtibial limb loss.

Selecting an optimal prosthetic foot is particularly challenging for highly active individuals with limb loss, such as military personnel, who need to seamlessly perform a variety of demanding activities/tasks (often with and without external loads) while minimizing risk of musculoskeletal injuries over the longer term. Here, we expand on prior work by comparing biomechanical and functional outcomes in two prosthetic feet with the largest differences in mechanical response to added load (i.e., consistently "Compliant" and "Stiff" forefoot properties). In each foot, fourteen male Servicemembers with unilateral transtibial limb loss (from trauma) completed instrumented gait analyses in all combinations of two loading conditions (with and without 22 kg weighted vest) and two walking speeds (1.34 and 1.52 m/s), as well as the Prosthesis Evaluation Questionnaire. With the Stiff foot, sound limb peak loading was 2% smaller (p = 0.043) in the loaded versus unloaded condition, but similar between loading conditions in the Compliant foot (note, the Stiff foot was associated with larger loads, overall). Independent of load or walking speed, the Compliant (versus Stiff) foot provided 67.9% larger (p < 0.001) prosthetic push-off, 17.7% larger (p = 0.01) roll-over shape radii, and was subjectively favored by 10 participants. A more Compliant versus Stiff prosthetic foot therefore appears to better accommodate walking with and without added load, and reinforce the notion that mechanical properties of prosthetic feet should be considered for near-term performance and longer-term (joint) health.

Associations between trunk postural control in walking and unstable sitting at various levels of task demand.

Trunk postural control (TPC) has been investigated in several populations and tasks. Previous work observed targeted training of TPC via isolated trunk control tasks may improve performance in other activities (e.g., walking). However, the nature of this relationship remains unknown. We therefore investigated the relationship between TPC, at both the global (i.e., response to finite perturbations) and local (i.e., resistance to continuous perturbations) levels, during walking and unstable sitting, both at varying levels of task demand. Thirteen individuals (11 Male, 2 Female) with no recent history (past 12 months) of illness, injury, or musculoskeletal disorders walked on a dual-belt treadmill at four speeds (-20%, -10%, +10%, and + 20% of self-selected walking speed) and completed an unstable sitting task at four levels of chair instability (100, 75, 60, and 45% of an individual's "neutral" stability as defined by the gravitational gradient). Three-dimensional trunk and pelvic kinematics were collected. Tri-planar Lyapunov exponents and sample entropy characterized local TPC. Global TPC was characterized by ranges of motion and, for seated trials, metrics derived from center-of-pressure time series (i.e., path length, 95% confidence ellipse area, mean velocity, and RMS position). No strong or significant correlations (-0.057 < ρ < 0.206) were observed between local TPC during walking and unstable sitting tasks. However, global TPC declined in both walking and unstable sitting as task demand increased, with a moderate inter-task relationship (0.336 < ρ < 0.544). While the mechanisms regulating local TPC are inherently different, global TPC may be similarly regulated across both tasks, supporting future translation of improvements in TPC between tasks.

Trunk and pelvic dynamics during transient turns among individuals with unilateral traumatic lower limb amputation.

Prior work has identified alterations in trunk-pelvic dynamics with lower limb amputation (LLA) during in-line walking; however, evaluations of other ambulatory tasks are limited. Turns are ubiquitous in daily life but can be challenging for individuals with LLA, prompting additional or unique proximal compensations when changing direction, which over time may lead to development of low back pain. We hypothesized such proximal kinematic differences between persons with and without LLA would exist in the sagittal and frontal planes. Three-dimensional trunk and pelvic kinematics, translational and rotational momenta, and coordination phase/variability were compared among eight persons with unilateral LLA (4 with transfemoral amputation and 4 with transtibial amputation), and five uninjured controls, who performed 90-degree turns to the left (n = 10) and right (n = 10). Participants self-selected the turn strategy (i.e., step vs. spin) and pivot limb in response to verbal cues regarding when and which direction to turn. Coordination variability and translational angular momenta did not differ between groups in either turn type. During spin turns, frontal rotational angular momenta were larger and frontal trunk-pelvis range of motion was smaller among persons with vs. without LLA. During step turns, pelvis leading transverse coordination was more frequent, frontal trunk rotational angular momentum was smaller, and sagittal pelvis range of motion was larger among persons with vs. without LLA. Altered and task-dependent modulation of trunk-pelvic dynamics among persons with LLA provides additional support for a potential link between repeated exposures to altered trunk-pelvic dynamics with elevated low back pain risk.

Modulation of Vertical Ground Reaction Impulse With Real-Time Biofeedback: A Feasibility Study.

Given its apparent representation of cumulative (vs peak) loads, this feasibility study investigates vertical ground reaction impulse (vGRI) as a real-time biofeedback variable for gait training aimed at reducing lower limb loading. Fifteen uninjured participants (mean age = 27 y) completed 12 2-min trials, 1 at each combination of 4 walking speeds (1.0, 1.2, 1.4, and 1.6 m/s) and 3 targeted reductions in vGRI (5, 10, and 15%) of the assigned ("target") limb, with the latter specified relative to an initial baseline (no feedback) condition at each speed. The ability to achieve targeted reductions was assessed using step-by-step errors between measured and targeted vGRI. Mean (SD) errors were 5.2% (3.7%); these were larger with faster walking speeds but consistent across reduction targets. Secondarily, we evaluated the strategy used to modulate reductions (ie, stance time or peak vertical ground reaction force [vGRF]) and the resultant influences on knee joint loading (external knee adduction moment [EKAM]). On the targeted limb, stance times decreased (P < .001) with increasing reduction target; first and second peaks in vGRF were similar (P > .104) across all target conditions. While these alterations did not significantly reduce EKAM on the target limb, future work in patients with knee pathologies is warranted.

A computational algorithm for classifying step and spin turns using pelvic center of mass trajectory and foot position.

Transient changes in direction during ambulation are typically performed using a step (outside) or spin (inside) turning strategy, often identified through subjective and time-consuming visual rating. Here, we present a computational, marker-based classification method utilizing pelvic center of mass (pCOM) trajectory and time-distance parameters to quantitatively identify turning strategy. Relative to visual evaluation by three independent raters, sensitivity, specificity, and overall accuracy of the pCOM-based classification method were evaluated for 90-degree turns performed by 3 separate populations (5 uninjured controls, 5 persons with transtibial amputation, and 5 persons with transfemoral amputation); each completed turns using two distinct cueing paradigms (i.e., laser-guided "freeform" and verbally-guided "forced" turns). Secondarily, we compared the pCOM-based turn classification method to adapted versions of two existing computational turn classifiers which utilize trunk and shank angular velocities (AV). Among 366 (of 486 total) turns with unanimous intra- and inter-rater agreement, the pCOM-based classification algorithm was 94.5% accurate, with 96.6% sensitivity (accuracy of spin turn classification), and 93.5% specificity (accuracy of step turn classification). The pCOM-based algorithm (vs. both AV-based methods) was more accurate (94.5% vs. 81.1-80.6%; P<0.001) overall, as well as specifically in freeform (92.9 vs. 80.4-76.8%; P<0.003) and forced (96.0 vs. 83.8-81.8%; P<0.001) cueing, and among individuals with (92.4 vs. 80.2-78.8%; P<0.001) and without (99.1 vs. 86.2-80.8%; P<0.001) amputation. The pCOM-based algorithm provides an efficient and objective method to accurately classify 90-degree turning strategies using optical motion capture in a laboratory setting, and may be extended to various cueing paradigms and/or populations with altered gait.