Background and purpose
Restoration of step length symmetry is a common rehabilitation goal after stroke. Persons post-stroke often retain the ability to walk with symmetric step lengths (“symmetric steps”); however, the resulting walking pattern remains effortful. Two key questions with direct implications for rehabilitation have emerged: 1) how do persons post-stroke generate symmetric steps, and 2) why do symmetric steps remain so effortful? Here, we aimed to understand how persons post-stroke generate symmetric steps and explored how the resulting gait pattern may relate to the metabolic cost of transport.
We recorded kinematic, kinetic, and metabolic data as nine persons post-stroke walked on an instrumented treadmill under two conditions: preferred walking and symmetric stepping (using visual feedback).
Gait kinematics and kinetics remained markedly asymmetric even when persons post-stroke improved step length symmetry. Impaired paretic propulsion and aberrant movement of the center of mass were evident during both preferred walking and symmetric stepping. These deficits contributed to diminished positive work performed by the paretic limb on the center of mass in both conditions. Within each condition, decreased positive paretic work correlated with increased metabolic cost of transport and decreased walking speed across participants.
It is critical to consider the mechanics used to restore symmetric steps when designing interventions to improve walking after stroke. Future research should consider the many dimensions of asymmetry in post-stroke gait, and additional within-participant manipulations of gait parameters are needed to improve our understanding of the elevated metabolic cost of walking after stroke.
Gait dysfunction is common after stroke . Persons post-stroke exhibit slow walking speeds [2,3,4], gait asymmetry [4, 5], and an elevated metabolic cost of transport (i.e., energy expended per meter walked) [6,7,8]. Gait training is a key component of stroke rehabilitation, as persons post-stroke frequently list gait improvement among their most desired rehabilitation goals .
Many rehabilitation approaches aim to restore step length symmetry [10,11,12,13,14,15,16]. The rationale for restoring step length symmetry is multifaceted: 1) asymmetric stepping increases the cost of transport in healthy adults , 2) persons post-stroke who walk with more asymmetric step lengths also tend to exhibit poorer balance  and more effortful gait patterns , 3) step length asymmetry is a simple metric that manifests from complex kinematic and kinetic asymmetries that can be difficult to treat in isolation, and 4) step length is easy to measure and manipulate in clinical settings (e.g., “step to the lines on the floor”). Consequently, there has been increasing interest in restoring step length symmetry after stroke, especially after recent intervention studies showed that improved step length symmetry coincided with improvements in gait speed  and cost of transport .
However, it is not clear that restoration of step length symmetry alone should lead to improvements in gait speed or cost of transport. Persons post-stroke often retain the capacity to walk with improved step length symmetry, even within a single testing session [16, 20, 21]. But unlike the intervention studies mentioned above, single-session studies have shown cost of transport to be similar whether persons post-stroke walk with asymmetric or symmetric step lengths [16, 21]. These findings suggest that improvements in gait speed and cost of transport likely arise from changes in kinematic or kinetic parameters that more directly influence gait speed or energetics and also affect step length symmetry. From this perspective, interventions that aim to restore step length symmetry but do not affect these critical underlying factors may not result in meaningful gait improvement. The ability to lessen cost of transport with an intervention aiming to restore step length symmetry likely depends on 1) the underlying causes of the asymmetry (which vary among patients [21, 22]), and 2) the mechanics used to generate the symmetric step lengths.
Here, we aimed to understand how persons post-stroke changed their walking patterns to restore step length symmetry and how these gait mechanics related to the cost of transport. We asked: do persons post-stroke restore step length symmetry by restoring symmetric gait mechanics or by relying on asymmetric compensatory mechanics? We hypothesized that persons post-stroke would restore step length symmetry using asymmetric walking patterns. We then aimed to explain why these asymmetric gait patterns cost so much energy despite improved step length symmetry.
Materials and methods
Ten persons post-stroke were recruited for the study. Data accrued from nine persons were retained for analysis (6 M/3F, age (mean ± SEM): 54 ± 4 years, lower extremity Fugl-Meyer : 26 ± 1, body mass: 93 ± 6 kg, all > 6 months post-stroke). Inclusion criteria for recruitment included a step length difference of at least 2 cm during over-ground walking. One participant was excluded from analysis because they unexpectedly reduced the asymmetry below 2 cm during treadmill walking. All other participants showed a > 2 cm step length difference during both over-ground and treadmill walking and reduced their step length asymmetry from the preferred walking trial to the symmetric stepping trial. Participants reported no additional neurological, musculoskeletal, or cardiovascular conditions. We determined preferred walking speed as the average speed of three over-ground 10-meter walk tests (0.81 ± 0.09 m/s, range: 0.40–1.25 m/s). Seven participants held onto the treadmill handrails, two wore ankle-foot orthoses, and one received functional electrical stimulation of the tibialis anterior. We asked participants who held onto the handrails to hold onto them as little as possible and avoid gripping the handrail if at all possible. All participants wore a safety harness that did not provide body weight support, provided written informed consent in accordance with the Johns Hopkins Medicine Institutional Review board prior to participation, and received monetary compensation.
We recorded kinematic (100 Hz) and kinetic (1000 Hz) data using a three-dimensional motion capture system (Vicon, Oxford, UK) and instrumented split-belt treadmill (Motek, Amsterdam, NL; Fig. 1a, left). We placed retroreflective markers over the seventh cervical vertebrae, tenth thoracic vertebrae, jugular notch, xiphoid process, and bilaterally over the second and fifth metatarsal heads, calcaneus, medial and lateral malleoli, shank, medial and lateral femoral epicondyles, thigh, greater trochanter, iliac crest, and anterior and posterior superior iliac spines. We filtered marker trajectories and ground reaction forces (GRFs) with fourth order low-pass Butterworth filters (6 Hz and 15 Hz cut-off frequencies, respectively). GRFs were set to zero for vertical GRF magnitudes < 32 N. Participants wore comfortable shoes and form-fitting clothing.