Differences in mediolateral dynamic stability during gait initiation according to whether the non-paretic or paretic leg is used as the leading limb.

We investigated mediolateral dynamic stability at first foot off and first initial contact during gait initiation according to whether the paretic or non-paretic leg was used as the leading limb. Thirty-eight individuals with stroke initiated gait with the paretic and non-paretic legs as the leading limb, and their movements were measured using a 3D motion analysis system. Margin of stability (i.e., the length between the extrapolated center of mass and lateral border of the stance foot) was used as an index of dynamic stability, with a large value indicating dynamic stability in the lateral direction. However, an excessively large margin of stability value (i.e., when the extrapolated center of mass is outside the medial border of the stance foot) indicates dynamic instability in the medial direction. Differences in the margin of stability between tasks were compared using the Wilcoxon signed-rank test. The minimum margin of stability was observed just before first foot off. When the non-paretic leg was used as the leading limb, the margin of stability tended to be excessively large at first foot off compared with when the paretic leg was used (p < 0.001). In other words, the extrapolated center of mass was outside the medial border of the paretic stance foot. In conclusion, lateral stability was achieved when using the non-paretic leading limb because the extrapolated center of mass was located outside the medial border of the stance foot. However, medial dynamic stability was lower for the non-paretic leading limb compared with the paretic leading limb.

Introduction

Posture is unconsciously controlled during steady-state walking but must be voluntarily controlled at gait initiation (GI) [1]. The cerebral cortex is activated before the floor reaction force changes in the GI task [2], and the basal ganglia, thalamus, and cerebellum are also involved at the start of movement [3]. Accordingly, various problems occur in patients with stroke, such as abnormal muscle activity [4] and delayed muscle activation [5]. The GI task is challenging in terms of the stability of control dynamics, particularly in the mediolateral direction [6]. In individuals with stroke, the paretic lower limb can be used as either the leading or trailing limb at GI. The leading limb is responsible for the initial transfer of weight to the stance limb as well as the antigravity flexion required for forward swing [7]. The trailing limb, on the other hand, is responsible for antigravity extension, which is required to support body weight and generate forward momentum for the first step [7]. Thus, physical therapists sometimes need to confirm that individuals have selected an effective leading limb, especially individuals with stroke.

Although evaluation of mediolateral stability is important for GI [8], few studies have analyzed differences in the selection of the leading limb from the viewpoint of stability. A recent systematic review argued against use of the paretic leading limb because it provides poor balance, noting also that the non-paretic leading limb stimulates the paretic postural muscles [4]. However, the same article also stated that use of the non-paretic leg as the leading limb poses a greater challenge in terms of balance during GI because the weakness of the paretic leg makes it difficult to support the body weight during the subsequent stance phase. A previous study comparing the change in the center of pressure (CoP), stride, and duration in 14 participants with stroke also recommended use of the paretic leading limb because the starting posture is asymmetrical and the first step length is shortened when the non-paretic leg is used as the leading limb [9]. The other previous study comparing ground reaction forces and GI speed in 13 participants with stroke recommended use of the non-paretic leading limb because, when subjects with stroke initiated with their non-paretic limb, the anteroposterior force and impulse generated by the trailing limb were strongly related to the magnitude of the final GI speed [10]. The reasons for these disagreements among the previous studies are as follows: i) few previous studies have investigated the selection of the leading limb, ii) different evaluation indices were used in the studies, and iii) the number of participants was small. The latter factor might have undermined the statistical power of the analyses.

Range of CoP movement is the classical method for analyzing postural stability [11,12]. The range of CoP movement is not a quantitative index for evaluating stability because the CoP does not move outside the base of support, even when balance is lost, and the range of the CoP movement is limited. A previous study [13] proposed the extrapolated center of mass (Xcom) as an index of dynamic motion stability based on use of an inverted pendulum model to calculate the margin of stability (MoS). In this study, the main outcome was MoS in the mediolateral direction, because mediolateral dynamic stability at first stance is the most important aspect for individuals with hemiplegia [14]. The aim of this study was to clarify the advantages and disadvantages of the use of the paretic and non-paretic legs as the leading limb from the viewpoint of mediolateral dynamic stability in individuals with stroke. Additionally, we sought to perform this study in a larger sample of patients than in previous studies. Previous work [9] has shown instability associated with the use of the non-paretic leading limb from the viewpoint of CoP movement. However, another study that analyzed the velocity of the center of mass indicated that use of the non-paretic leading limb was advantageous [10]. The leading limb also plays a role in predictive postural control [7], for which the non-paretic lower limb may be more effective. Therefore, here we examined the hypothesis that use of the non-paretic leg as the leading limb would be more stable from the viewpoint of dynamic stability in individuals with stroke compared with use of the non-paretic leg. Clarification of whether the choice of leading limb affects dynamic stability would provide information on the appropriate first step for individuals with stroke.

Materials and methods

Participants

A total of 38 individuals with stroke (mean age, 59.5 ± 10.1 years) participated in this cross-sectional study. Participant characteristics are shown in Table 1. All participants were patients in a convalescent rehabilitation ward who met the following inclusion criteria: (1) hemiparesis secondary to cerebrovascular accident; (2) first unilateral stroke; and (3) able to follow simple instructions and walk at least 10 m at their preferred speed without manual assistance. Individuals were excluded if they had other neurological or musculoskeletal deficits that would affect gait. This study was approved by the ethics committees of International University of Health & Welfare (17-Io-22) and Nakaizu Rehabilitation Center (28–007). Participants provided written informed consent before enrollment in the study.

Table 1

Participant characteristics (N = 38).

Age [years], mean ± SD	59.5 ± 10.1
Weight [kg], mean ± SD	59.2 ± 11.0
Height [cm], mean ± SD	163.9 ± 7.4
Sex (female/male)	11/27
Paretic side (right/left)	17/21
Time since onset (days), median (IQR)	92 (98)
BRS: lower extremity score, n (%)	III: 7 (18); IV: 9 (24); V: 18 (47); VI: 4 (11)
FMA: lower extremity score, median (IQR)	28 (3.75)
FMA: balance score, median (IQR)	10 (3.75)
FIM: transfer to the bed score, n (%)	V: 4 (13); VI: 18 (47); VII: 16 (42)
FIM: locomotion (walking) score, n (%)	IV: 4 (11); V: 10 (26); VI: 15 (39); VII: 9 (24)
10-m timed walk test (s), mean ± SD	20.7 ± 14.7

BRS, Brunnstrom recovery stage; FMA, Fugl–Meyer assessment; FIM, Functional Independence Measure; IQR, interquartile range.

Study protocol

During two tasks (i.e., use of the paretic leading limb and use of the non-paretic leading limb at GI), data were measured using a 3D motion capture system comprising eight Vicon MX cameras (Vicon Motion System Ltd., Oxford, UK) and six AMTI force plates (600 mm × 400 mm; Advanced Mechanical Technology Inc., Watertown, MA). Participants were positioned in a standardized static standing position with each foot placed on two separate force plates at pelvis width [15] because the wider the initial stance width, the greater the mediolateral movement [16]. Mediolateral stability changes according to the initial speed [8]. Thus, to ensure that the GI conditions were uniform for all participants, they were instructed to start walking as quickly as possible from a static standing position toward a line 3 m ahead [17] upon receiving a cue from a buzzer synchronized with an LED light (Fig 1). Following a reported protocol [9,18], participants wore normal low-heeled shoes; no ankle-foot orthosis or walking aid was used. First, participants repeated the GI trial three times with no instruction on which limb should be used as the leading limb; then, the participants were instructed to use the other leading limb for three trials. This procedure was chosen because random determination of the leading limb tended to cause the patients to freeze and behave unnaturally. Guarded assistance from physiotherapists during the task prevented the participants from falling.

Fig 1

Experimental set-up for GI with 34 reflective markers and 6 force plates.

A total of 34 reflective markers were attached to the participants at various landmarks, following a protocol used in previous studies [19,20]. The marker trajectories and force plate data were synchronized at a sampling frequency of 120 Hz. Marker trajectories and force plate data were low-pass filtered using a second-order Butterworth filter with cutoff values of 6 Hz and 18 Hz, respectively [21]. Center of mass and joint centers were calculated using anthropometric data [22] for the following 12 link-segment model: feet (head of the second metatarsal joint, fifth metatarsal joint, and heel markers), shanks (lateral malleolus, medial malleolus, and lateral femoral epicondyle markers), thighs (lateral femoral epicondyle, medial femoral epicondyle, and hip joint markers), pelvis (bilateral anterior superior iliac spine and posterior superior iliac spine markers), upper trunk (by T2 and T10 vertebrae, sternal notch, and jugular notch markers), upper arms (acromion and elbow lateral markers), and forearms (elbow lateral and wrist lateral markers). Joint kinematics and kinetics were calculated using an inverse dynamic model according to the Vicon Plug-in Gait model. Kinetic and kinematic data were exported into a biomechanics analysis program (Visual3D, version 5; C-Motion, Inc., Germantown, MD).

Parameters

The movement task was separated into three phases [23]: 1) the postural phase, from onset to first foot off; 2) the monopodal phase, from first foot off to first initial contact; and 3) the double support phase, from first initial contact to second foot off. Onset was defined as the start of the reverse reaction phenomenon of the CoP, namely, when the vertical component of the floor reaction force on the leading limb begins to continuously increase. To analyze the differences in dynamic stability at GI between each leading limb, we extracted the MoS, which is an index of dynamic motion stability at GI [8,24]. A previous study [13] proposed the use of Xcom, which is based on use of the inverted pendulum model to calculate the MoS. Using the formula from previous studies [13,25], the mediolateral MoS was calculated as the distance between Xcom and the boundaries of the base of support (BoS) during single stance from the foot markers, namely, the fifth metatarsal-phalangeal (MP) joint for the lateral border and the medial malleolus for the medial border. A visual representation of the MoS calculation is shown in Fig 2. The MoS can be calculated in medial and lateral directions and %MoS was calculated as the ratio of the distance (%) from the BoS boundary to Xcom, with the distance from the 5th MP to the medial malleolus taken as 100%.

Fig 2

Illustration of the MoS calculation at foot off.

The black circle indicates the location of Xcom while the white circles indicate key markers of the BoS boundary in the mediolateral direction. %MoS was calculated as the ratio of the distance (%) from the BoS boundary to Xcom, with the distance from the 5th MP to the medial malleolus being 100%. Abbreviations: 5th MP, fifth metatarsal-phalangeal joint; ω0, the eigenfrequency of the inverted pendulum model; CoM, center of mass; g, gravity; l, length of the leg; MoS, margin of stability; Vcom, velocity of the CoM; Xcom, extrapolated center of mass.

The values of %MoS were extracted at first foot off (%MoS-FO) and initial contact (%MoS-IC), as in a previous study [26]. A %MoS value lower than 0% indicates decreased dynamic stability in the lateral direction, while a larger %MoS indicates greater lateral stability. However, a %MoS value above 100% indicates that Xcom is inside the stance foot. Therefore, only a %MoS value between 0% and 100% would indicate stability in single support. Because stability in the lateral direction is the most important factor in patients with stroke [14], the primary index for determining stability is “MoS is a large value” and the secondary index is “MoS does not exceed 100%”.

The following additional indices were also used to compare the advantages and disadvantages of the leading limb selection, as in previous work [4,9]: movement duration, step length, and lateral pelvic tilt angle. Movement duration was calculated for each of three phases: 1) the postural phase, 2) monopodal phase, and 3) double support phase. In previous studies, these indicators were used to determine the advantage of the leading limb selection from the viewpoint of movement efficiency (duration) or posture symmetry (step length or lateral pelvic tilt). Step length was calculated as the anteroposterior distance between the heel markers at initial contacts. The lateral pelvic tilt was calculated based on changes from the static standing position to first foot off because the pelvis is lifted most at first foot off in individuals with stroke [27].

GI tasks were measured without the use of walking aids, even in those who usually use canes (17 persons) and orthoses (13 persons). Accordingly, in patients who use walking aids, the GI movement might have a different pattern. Therefore, as an additional analysis, the participants were divided into two groups (walking aid user or not), and the differences in %MoS-FO and %MoS-IC were compared in each group between the tasks performed using paretic and non-paretic leading limbs.

Statistical analysis

Differences in %MoS, movement duration, step length, and lateral pelvic tilt were compared between the tasks performed using the paretic and non-paretic leading limbs. All indices were extracted as the average of three trials for each task. The Shapiro–Wilk test was used to assess the normality of all data. Normally distributed data were compared using the paired t-test, whereas non-normally distributed data were compared using the Wilcoxon signed-rank test (SPSS, version 24.0; IBM Corp., Armonk, NY). The significance level was set at α = 0.05, and the effect size r [28] was calculated for each index. According to Cohen [29], a large effect is represented by an r of at least 0.50, a moderate effect by 0.30, and a small effect by 0.10. The formula for calculating the effect size r was where Z is the Z score on the Wilcoxon signed-rank test and n is the number of pairs.

Results

In the first three trials, 17 participants used the paretic leg consecutively as the leading limb and 6 used the non-paretic leg. Fifteen participants did not select a specific leading limb in consecutive trials. The differences in the average %MoS displacements in the mediolateral direction are shown in Fig 3. No participants had negative %MoS values in the mediolateral direction. Regardless of which limb was used as the leading limb at GI, %MoS showed its minimum value just before the first foot off, namely, Xcom moved the most in the lateral direction at this time. In addition, in most participants, %MoS-FO tended to be from 0% to 100% when the paretic leg was used as the leading limb. When the non-paretic leg was used as the leading limb (solid line), Xcom tended to move less toward the paretic foot and the absolute minimum %MoS value exceeded 100% in 20 of the 38 participants, that is, Xcom did not reach the BoS comprising the paretic lower limb. When the paretic leg was used as the leading limb (dotted line), the absolute minimum %MoS value exceeded 100% in only 8 of the 38 participants.

Fig 3

Differences in the average %MoS between tasks performed using paretic and non-paretic leading limbs (n = 38).

The solid line indicates the task performed using the non-paretic leading limb. The dotted line indicates the task performed using the paretic leading limb. Dashed lines around the dotted line for the paretic leg and fine solid lines for the non-paretic leg indicate the standard deviation. The trajectories were obtained for the averaged trial of each subject with GI begun as quickly as possible from a static standing position. The horizontal axis is time (%), which is normalized to 100% each from onset to first foot off and from first foot off to first initial contact. The vertical axis is the distance (%) from the lateral boundary of the BoS to Xcom. The illustration of the foot is an image of the mediolateral position of the first stance foot. Mediolateral BoS (5th MP to medial malleolus) is normalized to 100, with negative values indicating instability in the lateral direction. Abbreviations: 5th MP, fifth metatarsal-phalangeal joint; BoS, base of support; CoM, center of mass; FO, foot off; IC, initial contact; MoS, margin of stability; Xcom, extrapolated center of mass.

Differences in the mediolateral %MoS, movement duration, step length, and lateral pelvic tilt angle are shown in Table 2. The results showed significant differences in %MoS-FO and %MoS-IC with a large effect size (r = 0.82, P < 0.001). The median difference between tasks was 24%. When the non-paretic leg was used as the leading limb, %MoS-FO and %MoS-IC were larger compared with when the paretic leading limb was used. Regarding the movement duration, when the non-paretic leading limb was used, the postural phase was longer and the monopodal support phase was 0.13 seconds shorter than when the paretic leading limb was used (effect size r = −0.66, P < 0.001). The pelvic lateral tilt angle reached its maximum value at first foot off. When the non-paretic leg was used as the leading limb, the lateral pelvic tilt was 2.8° smaller at first foot off compared with when the paretic leading limb was used (effect size r = 0.76, P < 0.001). In additional analysis that divided the participants into two groups (walking aid user or not), %MoS-FO and %MoS-IC showed similar results and there were significant differences only in walking aid users (n = 20). The non-walking aid users (n = 18) had no significant differences in %MoS-FO between use of the paretic and non-paretic leading limb. However, they had the same tendency in the above result; that is, %MoS-FO exceeded 100% when a non-paretic leg was used as the leading limb.

Table 2

Differences between use of the paretic and non-paretic leg as the leading limb (N = 38).

	Paretic leading limb	Non-paretic leading limb	p-value	Effect size (r)
%MoS (%)
%MoS-FO#, median (IQR)	86.1 (22.8)	110.8 (34.5)	<0.001	0.82
%MoS-IC#, median (IQR)	155.0 (28.0)	172.8 (50.1)	0.008	0.50
Duration (s)
Postural phase#, median (IQR)	0.18 (0.10)	0.20 (0.18)	0.014	0.40
Monopodal phase#, median (IQR)	0.43 (0.19)	0.30 (0.10)	<0.001	−0.66
Double support phase#, median (IQR)	0.28 (0.22)	0.27 (0.20)	0.294	0.17
Step length (cm/Ht), mean ± SD
First step	0.199 ± 0.077	0.208 ± 0.077	0.290	0.17
Second step	0.182 ± 0.095	0.189 ± 0.081	0.419	0.13
Lateral pelvic tilt angle at first FO (deg)#, median (IQR)	4.14 (3.30)	1.34 (4.51)	<0.001	0.76

#Non-normally distributed; %MoS-FO, %MoS at first foot off; %MoS-IC, %MoS at first initial contact; Ht, height; IQR, interquartile range; MoS, margin of stability; postural phase, onset to first foot off; monopodal phase, first foot off to first initial contact; and double support phase, first initial contact to second foot off.

Discussion

This is the first study to clarify which leading limb is more stable from the viewpoint of mediolateral dynamic stability in individuals with stroke. When the non-paretic leg was used as the leading limb, %MoS-FO and %MoS-IC were larger in the medial direction at GI compared with when the paretic leading limb was used. This result indicates dynamic stability in the lateral direction with the non-paretic leading limb. However, regarding the medial direction, this result was contrary to our hypothesis that use of the non-paretic leg as the leading limb is more stable for individuals with stroke for GI than use of the paretic leg as the leading limb.

GI poses a challenge in terms of mediolateral dynamic stability in individuals with stroke [4,8,18] because they have various problems associated with the transfer of weight onto the paretic limb, such as sensory loss, muscle weakness, and fear. Individuals with stroke tend to adopt an asymmetric posture by reducing the load on the paretic limb not only during static standing, but also for steps and when walking [9,30,31]. In this study, the monopodal phase was shortened when the non-paretic leading limb was used. This is also one of the compensatory strategies to reduce loading on the paretic limb. Other previous studies of GI have also mentioned this tendency after stroke in the subacute phase [9] and chronic phase [10]. In this study, %MoS-FO and %MoS-IC exceeded 100% in more than half of the participants when the non-paretic limb was used as the leading limb. Thus, these negative behaviors toward paretic limb loading may have caused an excessively large MoS when the non-paretic leading limb was used, that is, when the paretic leg was used as the first stance limb.

In a previous study comparing MoS in the lateral direction and various balance indices in individuals with stroke, clinical balance indices had a negative correlation with MoS in the lateral direction of the paretic leg [32]. This previous research mentioned that maintenance of a large MoS in the lateral direction indicated a poor ability to maintain balance. Therefore, a large MoS in the paretic direction indicates that they might intentionally limit the center of mass transfer on the paretic side. This compensatory movement to focus on instantaneous balance is the reason why %MoS exceeded 100% when the non-paretic leg was used as the leading limb (i.e., as the first stance phase of the paretic limb).

In single support, only a %MoS value between 0% and 100% indicates stability. Therefore, especially for dynamic stability in the lateral direction, we conclude that use of the non-paretic leading limb results in higher dynamic stability than use of the paretic leading limb. However, especially for dynamic stability in the medial direction, we conclude that use of the non-paretic leading limb results in worse dynamic stability than use of the paretic leading limb. Patients with stroke may intentionally maintain such an excessive %MoS to avoid lateral perturbation, which leads to falls. The excessive lateral stability might not lead to a feeling of reassurance in individuals after stroke, even though the kinematic stability of the lateral direction was guaranteed, because the fact that the MoS persistently exceeded 100% suggests instability in the medial direction, with participants forced to immediately take another step. Some previous studies reported that adults and children with hemiplegia tend to use a paretic leading limb because of their asymmetric standing posture, which is the result of less limb loading on the paretic side [30,33]. The postural phase was shortened in our study, with the first foot off earlier with the paretic leading limb than with the non-paretic leading limb. This result may indicate that individuals with stroke may find it easier to lift the paretic lower limb as the leading limb.

Regarding other indices, there was no difference in step length, regardless of the leading limb selection, which is in contrast to the result of a previous study [9]. However, there was a significant difference in the lateral pelvic tilt at first foot off, which showed a large value when the paretic leg was used as the leading limb. This result is similar to that of Davies et al. [27], who found that, if patients start walking using the paretic leading limb, the pelvic is excessively actively lifted. This seems to be compensatory hip hiking for toe clearance [34]. The pelvic tilt angle when the non-affected leading limb was used (median, 1.34°) was similar to that in healthy individuals (average, 0 ± 1°) [35]. However, the angle when the paretic leading limb was used (median, 4.14°) was three times larger than that when the non-paretic leading limb was used. The minimal clinically important difference (MCID) for the pelvic elevation angle has not been determined. Nonetheless, it can be said that GI deviates asymmetrically from normal movement when the paretic leg is used as the leading limb.

Given the above, we can conclude that use of the non-paretic leading limb is disadvantageous for mediolateral stability because %MoS-FO and %MoS-IC values exceeding 100% may indicate instability, unless the patient steps rapidly in the medial direction. However, there is an advantage in postural symmetry. This insufficient weight transfer to the paretic limb can represent compensatory movement to obtain biomechanical stability when the non-paretic leading limb is used. Instruction of individuals with stroke to use the paretic leg as the leading limb can be useful from the viewpoint of mediolateral dynamic stability. However, it may be necessary to suggest use of the non-paretic limb as the leading limb to individuals with stroke and high balance function to improve postural symmetry because the non-walking aid users had no significant differences in %MoS-FO between use of the paretic and non-paretic leading limbs.

There are several limitations to this study. First, because it was conducted with limited evaluation indices, it was not possible to analyze the results based on the participants’ physical function, use of a walking aid, dominant foot, and favored leading limb. Second, we focused on only kinematic indicators, and other factors related to subjective impressions may also need to be considered regarding leading limb selection. Third, to eliminate the effect of walking aids, GI tasks were measured without the use of walking aids, even in those who usually use canes (17 persons) and orthoses (13 persons). Accordingly, if the patients use walking aids, the GI movement might have different patterns. Despite these limitations, few other studies have examined the selection of the leading limb at the initiation of walking in a large patient sample and the results of this study could thus be useful for the development of rehabilitation programs focusing on movements at walking initiation.

Conclusions

This study examined the mediolateral dynamic stability in GI from the viewpoint of differences in the selection of the leading limb in individuals with stroke. When poststroke individuals used the non-paretic leg as the leading limb, the MoS in the lateral direction was excessively large compared with when the paretic limb was used. Therefore, medial dynamic stability was lower when individuals with stroke initiated walking with the non-paretic leg than with the paretic leg. These results should help physical therapists when they are recommending the appropriate first step for individuals with stroke.

(XLSX)

Click here for additional data file.

13 Jan 2022

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Peter Andreas Federolf

Academic Editor

PLOS ONE

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. We note that you have indicated that data from this study are available upon request. PLOS only allows data to be available upon request if there are legal or ethical restrictions on sharing data publicly. For more information on unacceptable data access restrictions, please see http://journals.plos.org/plosone/s/data-availability#loc-unacceptable-data-access-restrictions.

In your revised cover letter, please address the following prompts:

a) If there are ethical or legal restrictions on sharing a de-identified data set, please explain them in detail (e.g., data contain potentially sensitive information, data are owned by a third-party organization, etc.) and who has imposed them (e.g., an ethics committee). Please also provide contact information for a data access committee, ethics committee, or other institutional body to which data requests may be sent.

b) If there are no restrictions, please upload the minimal anonymized data set necessary to replicate your study findings as either Supporting Information files or to a stable, public repository and provide us with the relevant URLs, DOIs, or accession numbers. For a list of acceptable repositories, please see http://journals.plos.org/plosone/s/data-availability#loc-recommended-repositories.

We will update your Data Availability statement on your behalf to reflect the information you provide.

3. Please amend either the abstract on the online submission form (via Edit Submission) or the abstract in the manuscript so that they are identical.

4. Please include your full ethics statement in the ‘Methods’ section of your manuscript file. In your statement, please include the full name of the IRB or ethics committee who approved or waived your study, as well as whether or not you obtained informed written or verbal consent. If consent was waived for your study, please include this information in your statement as well.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Partly

Reviewer #2: Partly

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: No

Reviewer #2: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: This study investigated the mechanics of gait initiation among patients post stroke, comparing their gait initiation with the paretic vs their non-paretic leg. The study is highly relevant for stroke rehabilitation in practice and adds value to understanding gait mechanisms post-stroke. However, there are certain methodological issues that will need to be addressed before being eligible for publication.

The main issue with this paper is the interpretation of the results leading to the main conclusion; differences are found in %MoS during GI, with a lower %MoS when the paretic leg is leading. This finding led to the conclusion that using the non-paretic leg as the leading limb is more stable compared to using the paretic leg, however, alternative conclusions are possible from this finding. That is, when the paretic leg led, movement times in the monopodal phase were significantly longer and in single support the interpretation of a higher %MoS as an indicator for stability is void. That is, in single support only a %MoS between 0 and 100 would indicate stability. The decreased values of %MoS could in this case indicate increased stability or a movement strategy more focused on instantaneous balance rather than quick progression.

A second major issue is that there is currently insufficient detail in the methods to reproduce the study. Especially in terms of the computation of the dependent variables, this should be described better. For instance, the following issues should be clarified:

• The protocol was performed three times per leading leg. How were these three trials analyzed, averaged or on a trial-by-trial level?

• The exact timing of sampling %MoS is unclear, for this, it would help to introduce the concepts of the paragraph on lines 127-135 earlier in the methods, perhaps supported by a figure.

• Related to the above point, in the results it is mentioned that “the %MoS showed its minimum value at first foot off” (line 152-153), whereas figure 3 shows this minimum to be just prior to first foot off.

• Finally related to this, in the results and discussion, the authors often mention %MoS without indicating which sampling moment is meant. I would suggest specifying with as %MoS-FO, %MoS-IC, or %MoS-both (or using subscript)

A final methodological issue lies in one of the factors that the authors also highlighted as a limitation; namely that all participants have been analyzed walking without aid, even though they might not be used to this. This analysis might not be sufficiently powered to lead to strong conclusions, but it would be helpful to add a subgroup analysis to get an indication of whether the main results are similar across different groups or driven by one group in particular.

Further minor issues need to be considered:

• The abstract is confusing when read without prior knowledge of the paper. For instance, terms as ‘first foot off’ should be more clearly described as well as a % margin of stability of over 100% (to be normalized to 100% implies that 100% is the maximum value), also it is not mentioned in which direction (medial or lateral) these values are directed.

• Line 39, posture is not always unconsciously controlled during walking and while it is true that gait initiation needs to be done voluntary, it could be debated to what extend posture is consciously controlled in this task.

• Line 45-47. “Thus, physical …” to “… with stroke.” This statement does not flow well. Why does a physical therapist determine this? Should patient not decide this?

• Line 59-60. Displacement needs to be defined better. Displacement cannot be defined as instability as the CoP is always moving, without always being instable. Furthermore, this neglects the functional role of CoP fluctuations (e.g. van Emmerik; van Wegen, doi:10.1097/00003677-200210000-00007).

• Line 66-67 & 75. Why was the number of 38 individuals chosen? Was a power calculation done based on the mentioned (but underpowered) previous studies?

• Line 85-107. How was a loss of balance prevented in this study? Did the participants wear a harness or was there a therapist nearby?

• Line 154-158. When discussing for how many participants %MoS exceeded 100%, was this measured on the two instances %MoS was sampled or the absolute minimal %MoS value for a participant?

• Line 158. “No participants had negative %MoS values in the mediolateral direction.” A negative %MoS would very quickly lead to a loss of stability and/or fall, which I assume would be prevented in the experimental design, to avoid injuries to the participant. Was it even possible to score a negative %MoS?

• Line 173. No hypothesis has been stated in the introduction.

• Figure 3. The standard deviation lines are not quite clear because of the added shading and their overlap. I would suggest taking away the shading and using a dotted line around the dashed line for the paretic leg and a fine solid line to indicate the standard deviation for the non-paretic leg.

Finally, I have highlighted elsewhere in the review system that I believe more could be done to adhere to the Data Availability policy of PLOS. I would like to further clarify this point here. I believe this is a very relevant study which collected data in a complex setting from a very select group. Sharing the raw data would not only increase the value of this article (improve reproducibility and allow for external validation), it would also hold ethical benefits as it would allow others to perform pilot analyses, without using valuable time and effort of this vulnerable patient group.

Reviewer #2: The data in the manuscript partly support the conclusions; however, the authors appear to overreach in the conclusions drawn from the margin of stability metric, which, although is a good measure of mediolateral postural stability, does not fully capture an individual's postural control during a task. Please see a complete list of detailed comments in the attached document.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

Submitted filename: PONE-D-21-29688_Review.pdf

Click here for additional data file.

10 Feb 2022

Detailed Response to Reviewers.

Reviewer #1:

We wish to express our deep appreciation for your feedback. As indicated in our point-by-point responses below, we have taken all of your comments and suggestions into consideration in the revised manuscript. Major revisions are highlighted in yellow in the marked-up copy of the manuscript.

Comments

The main issue with this paper is the interpretation of the results leading to the main conclusion; differences are found in %MoS during GI, with a lower %MoS when the paretic leg is leading. This finding led to the conclusion that using the non-paretic leg as the leading limb is more stable compared to using the paretic leg, however, alternative conclusions are possible from this finding. That is, when the paretic leg led, movement times in the monopodal phase were significantly longer and in single support the interpretation of a higher %MoS as an indicator for stability is void. That is, in single support only a %MoS between 0 and 100 would indicate stability. The decreased values of %MoS could in this case indicate increased stability or a movement strategy more focused on instantaneous balance rather than quick progression.

Response:

Thank you for your feedback. In response to your advice, we have changed the interpretation of the results from “higher %MoS value indicate good stability” to “a %MoS exceeding 100% suggests instability.” Therefore, the conclusion was changed to “use of the non-paretic leading limb results in worse dynamic stability than use of the paretic leading limb. Patients with stroke may intentionally maintain such an excessive %MoS to avoid lateral perturbation, which leads to falls.” (lines 33–38, 290–293, 315–322, 346–357, 375–377).

A second major issue is that there is currently insufficient detail in the methods to reproduce the study. Especially in terms of the computation of the dependent variables, this should be described better. For instance, the following issues should be clarified:

• The protocol was performed three times per leading leg. How were these three trials analyzed, averaged or on a trial-by-trial level?

• The exact timing of sampling %MoS is unclear, for this, it would help to introduce the concepts of the paragraph on lines 127-135 earlier in the methods, perhaps supported by a figure.

• Related to the above point, in the results it is mentioned that “the %MoS showed its minimum value at first foot off” (line 152-153), whereas figure 3 shows this minimum to be just prior to first foot off.

• Finally related to this, in the results and discussion, the authors often mention %MoS without indicating which sampling moment is meant. I would suggest specifying with as %MoS-FO, %MoS-IC, or %MoS-both (or using subscript)

Response:

Thank you for your feedback. All indices were extracted for each task and then the average of three trials was used (line 217).

Regarding the exact timing of the %MoS sampling, we have defined the phase division of the movement task at the beginning of the Parameters section (lines 159–164). We extracted the feature values according to this phase classification.

The minimum %MoS value was found just before the first foot off. Accordingly, we have changed the sentence from “the %MoS showed its minimum value at the first foot off” to “the %MoS showed its minimum value just before the first foot off” (line 234). Additionally, the sampling moment was specified by using %MoS-FO and %MoS-IC, in response to your suggestion.

A final methodological issue lies in one of the factors that the authors also highlighted as a limitation; namely that all participants have been analyzed walking without aid, even though they might not be used to this. This analysis might not be sufficiently powered to lead to strong conclusions, but it would be helpful to add a subgroup analysis to get an indication of whether the main results are similar across different groups or driven by one group in particular.

Response:

Thank you for your feedback. In response to your advice, we divided the participants into two groups (walking aid user or not) and reanalyzed the %MoS (lines 207–212). Both groups had the same tendency, and there were significant differences only in walking aid users between when the paretic leg led and the non-paretic leg led. Thus, we have added this finding to the Results (lines 270–276).

Further minor issues need to be considered:

• The abstract is confusing when read without prior knowledge of the paper. For instance, terms as ‘first foot off’ should be more clearly described as well as a % margin of stability of over 100% (to be normalized to 100% implies that 100% is the maximum value), also it is not mentioned in which direction (medial or lateral) these values are directed.

Response:

Thank you for your suggestion. We have clarified that we investigated mediolateral dynamic stability by focusing on first foot off and the first initial contact of the leading limb in the Abstract (lines 20–21). Additionally, the direction of the positive margin of the stability value was specified (lines 26–29).

• Line 39, posture is not always unconsciously controlled during walking and while it is true that gait initiation needs to be done voluntary, it could be debated to what extend posture is consciously controlled in this task.

Response:

Thank you for your suggestion. We have added other previous studies indicating that brain activity precedes GI and argue that there are changes in muscle activity in GI in patients with stroke (lines 42–46).

• Line 45-47. “Thus, physical …” to “… with stroke.” This statement does not flow well. Why does a physical therapist determine this? Should patient not decide this?

Response:

Thank you for your suggestion. We have changed the sentence to “physical therapists sometimes need to confirm that individuals have selected an effective leading limb, especially individuals with stroke.” (lines 53–54).

• Line 59-60. Displacement needs to be defined better. Displacement cannot be defined as instability as the CoP is always moving, without always being instable. Furthermore, this neglects the functional role of CoP fluctuations (e.g. van Emmerik; van Wegen, doi:10.1097/00003677-200210000-00007).

Response:

Thank you for your feedback. As you pointed out, “displacement of CoP” is an ambiguous definition. Accordingly, we have changed the term to “range of CoP movement” and defined it as the “peak to peak amplitude of the CoP movement” (lines 75–79).

• Line 66-67 & 75. Why was the number of 38 individuals chosen? Was a power calculation done based on the mentioned (but underpowered) previous studies?

Response:

Thank you for your question. Because no previous study has analyzed the MoS during GI in individuals with stroke, which is the main outcome in this study, we could not estimate the sample size.

• Line 85-107. How was a loss of balance prevented in this study? Did the participants wear a harness or was there a therapist nearby?

Response:

Thank you for your suggestion. Guarded assistance during the task prevented the participants from falling. The participants did not wear a harness in this study. We have added this information to the Study protocol section (lines 135–136).

• Line 154-158. When discussing for how many participants %MoS exceeded 100%, was this measured on the two instances %MoS was sampled or the absolute minimal %MoS value for a participant?

Response:

Thank you for your suggestion. The number of participants who had a %MoS that exceeded 100% was determined by the absolute minimum %MoS value. We have added the explanation “the absolute minimum %MoS value exceeded 100% in 20 of the 38 participants” (lines 239 and 241–242).

• Line 158. “No participants had negative %MoS values in the mediolateral direction.” A negative %MoS would very quickly lead to a loss of stability and/or fall, which I assume would be prevented in the experimental design, to avoid injuries to the participant. Was it even possible to score a negative %MoS?

Response:

Thank you for your suggestion. We previously identified a negative MoS in the lateral direction during gait measurement, such as near-falls (Osada et al.; doi: 10.1016/j.arrct.2021.100156). The participants were stopped by manual assistance to prevent falling immediately after a negative MoS value. Because there were no such near-falls in this study, we stated that “No participants had negative %MoS values in the mediolateral direction.”

• Line 173. No hypothesis has been stated in the introduction.

Response:

Thank you for your suggestion. Our hypothesis has been added to the end of the Introduction (lines 92–96).

• Figure 3. The standard deviation lines are not quite clear because of the added shading and their overlap. I would suggest taking away the shading and using a dotted line around the dashed line for the paretic leg and a fine solid line to indicate the standard deviation for the non-paretic leg.

Response:

We agree with this comment. We have modified Figure 3 in line with your suggestion.

Finally, I have highlighted elsewhere in the review system that I believe more could be done to adhere to the Data Availability policy of PLOS. I would like to further clarify this point here. I believe this is a very relevant study which collected data in a complex setting from a very select group. Sharing the raw data would not only increase the value of this article (improve reproducibility and allow for external validation), it would also hold ethical benefits as it would allow others to perform pilot analyses, without using valuable time and effort of this vulnerable patient group.

Response:

Thank you for your suggestion. We have added all relevant data within the paper and its Supporting Information files.

Reviewer #2:

We wish to express our deep appreciation for your feedback. As indicated in our point-by-point responses below, we have taken all of your comments and suggestions into consideration in the revised manuscript. Major revisions are highlighted in yellow in the marked-up copy of the manuscript.

General Comments

1. The authors’ intended meaning of dynamic stability is not clearly defined and I believe is a point of confusion throughout the manuscript as they also mention balance control, particularly in the discussion. Being in a stable position is not the same as being able to control that position. Whether the paretic or non-paretic limb should be used to initiate gait would be dependent on a combination of these factors: in which configuration (paretic vs. non-paretic leading limb) are they most stable and in which configuration do they have the best ability to control their balance? This important distinction and consideration is not addressed by the authors, which I believe leads to the contradictions noted in the discussion (see next comment).

Response:

Thank you for your suggestion. We have clearly defined dynamic stability as “only a %MoS between 0 and 100 would indicate stability in single support” (lines 192–195) and modified the Discussion (lines 290–293, 315–322, 346–349).

2. In general, the discussion tends to contradict the authors’ conclusion more than it supports their conclusion (see specific comments 16-18). This is a major concern which suggests that the authors are not drawing appropriate conclusions from their data.

Response:

Thank you for your suggestion. In response to your specific comments 16–18, we have improved the consistency of the Discussion by modifying the interpretation of the %MoS.

3. The methods mention the subjects had a favored leading limb and that this limb was always used first in the task. However, it is not clarified whether participants “favored” their paretic or non-paretic limb and how this might have affected their findings. How did results differ between the favored and unfavored limb as compared to the paretic and non-paretic limb? Which limb did the participants select and how did this compare to their biomechanics based on your metrics?

Response:

Thank you for your question. Of the 38 participants, 17 favored the use of the paretic leg as the leading limb. We have added this number to the beginning of the Results section (line 229). However, we believe that the article will be complicated and confusing if we add an analysis of the preferences of the leading limbs. Therefore, we have stated that it is necessary to analyze the biomechanics based on the leading limb preference in the Study limitations (lines 360–361).

4. Please put citations inside punctuation. This greatly improves readability of the manuscript.

Response:

Thank you for your suggestion. We have placed the citations inside punctuation.

Specific Comments

Introduction

1. Line 39: The authors state that posture is unconsciously controlled during walking; however, it would be more accurate to state that posture is unconsciously controlled during “steady-state” or “unperturbed” walking.

Response:

Thank you for your suggestion. We have changed “Posture is unconsciously controlled during walking” to “Posture is unconsciously controlled during steady-state walking” (line 41).

2. Lines 49-51: The interpretation of the findings of reference 5 seem to be overstated by the authors. While the article does suggest that the use of the paretic leg as the leading leg is associated with difficulties activating the tibialis anterior and gluteus medius muscles during the anticipatory postural adjustment phase of gait initiation, the same article also states “In contrast, the use of the paretic leg as the trailing leg seems to challenge balance to a greater extent during GI” and in their conclusion “GI is facilitated when the non-paretic leg is used as the trailing leg because the weakness of the paretic leg leads to difficulties in supporting body weight during the upcoming stance phase.” Please accurately present the findings of the previous study in the introduction.

Response:

To address this comment, we have added the following findings of the previous study to the Introduction: “However, the same article also states that use of the non-paretic leg as the leading limb seems to challenge balance to a greater extent during GI because the weakness of the paretic leg leads to difficulties in supporting body weight during the upcoming stance phase.” (lines 59–62).

3. Lines 54-57: This sentence requires some clarification. First, what were the different evaluation indices used in each study? Second, it is unclear what the conclusions of references 7 and 8 (Tokuno and Brunt, respectively) found with regards to which leading limb is preferable.

Response:

To address this comment, we have added a clarification of each previous study. Hesse et al. compared the change in CoP, stride, and duration in 14 participants with stroke and recommended use of the paretic leading limb. Tokuno and Eng compared the ground reaction forces and GI speed in 13 participants with stroke and recommended use of the non-paretic leading limb. A previous study by Brunt et al. has been removed because it does not mention the superiority of the leading limb selection during GI (lines 63–70).

4. Line 61: Statement should read “However, center of pressure displacement…”

Response:

Thank you for this suggestion. We have added “displacement” to the sentence (line 77).

5. Lines 62-64: Please elaborate on why the use of the extrapolated center of mass/margin of stability is a superior measure of dynamic balance than center of pressure displacement.

Response:

Thank you for raising this point. We have added an explanation of why the use of center of pressure is not better than that of extrapolated center of mass (lines 77–81).

6. Lines 67-69: Please more clearly define the purpose statement. As written, the purpose is unnecessarily wordy.

Response:

Thank you for your suggestion. We have clarified the purpose of this study (lines 88–90).

Methods

7. Lines 82-83: Please explicitly state the university and hospital which gave approval for this study. Also, this sentence does not need to be its own paragraph.

Response:

Thank you for your suggestion. We have added the university and hospital name (lines 109–110).

8. Lines 96-98: Several issues need to be addressed: What is meant by “favored” leading limb and how was this determined? Did the participant’s preference of favored limb influence any of the outcome measures? What are the potential limitations of not randomizing the initial leading limb? Was the initial stance width controlled? How might the initial stance width influence the MOS?

Response:

Thank you for your suggestion. We have added the reason why the leading limb was not determined randomly (lines 133–135). Eventually, we obtained three trials for each task, without randomization of the leading limb selection. This is important because random determination of the leading limb tended to cause the patient to freeze and behave unnaturally. Therefore, first, participants repeated the trial without an instruction on which limb to be used as the leading limb, and then, participants were instructed to use the other leading limb for three more trials.

The initial stance width was controlled with the pelvis width because the wider the initial stance width, the greater the mediolateral movement (line 124).

9. Lines 103-104: Given the importance of accurately identifying the center of mass location in your primary outcome measure, the calculation of the center of mass location should be more fully explained.

Response:

Thank you for your suggestion. We have added the details of how to calculate the center of mass. According to a previous study (Dempster, 1955), center of mass was calculated using anthropometric data for the following 12 link-segment model: foot (2), lower leg (2), thigh (2), pelvis, upper trunk, upper arm (2), and lower arm (2) (lines 145–153).

10. Lines 127-128: What is the purpose of the “additional indices” used in the study? Why were these three selected? What additional information do they provide?

Response:

Thank you for your suggestion. The additional indices were used to compare the advantages and disadvantages of leading limb selection with the results of a previous study. In each previous study, these indicators were used to state the advantage of the leading limb selection (lines 196–197, 200–202).

11. Lines 128-132: It would be helpful to number the 3 phases (e.g., 1) the postural phase, …; 2) the monopodal phase…).

Response:

We agree with your suggestion. We have added the number to the three phases (lines 159–161, 199–200).

12. Lines 134-135: The calculation of the lateral pelvic tilt angle is not sufficiently explained. What markers were used?

Response:

Thank you for your suggestion. We have added the details of the attached marker points (lines 149–150).

13. In Figure 3, the time is defined from 0-200%; however, nowhere in the methods is the time normalization of the data explained. Please define these methods.

Response:

Thank you for this suggestion. We have changed Figure 3 from 200% normalized to 100% normalized to match the figure legends.

Results

13. While p-values/effect sizes are presented in Table 2, it would be helpful to have these data included in the text. Moreover, the authors should provide a metric of how much larger/smaller (raw difference or percent difference between averages/medians) a variable was for a given condition when describing their results.

Response:

Thank you for your suggestion. We have added the p-values, effect size, and raw difference between medians to the Results section (lines 261, 267, 270).

Discussion

14. Lines 183-184: In which condition did the %MoS exceed 100% in more than half of the participants? Please clarify.

Response:

Thank you for your question. We have clarified the conditions under which the %MoS exceeded 100% (lines 303–304).

15. Lines 184-186: The statement about negative behaviors associated with paretic limb loading seem to contradict the authors’ assertion that the non-paretic leading limb (paretic trailing-limb) is better.

16. Lines 188-190: The statement that a large MOS in the lateral direction indicated poor ability to maintain balance appears to contradict the author’s assertion that the non-paretic leading limb is better based on their findings that a larger MOS was observed with the non-paretic leading limb.

17. Lines 203-204: Again, this statement that individuals with stroke may find it easier to used the paretic limb as the leading limb contradicts their conclusion.

Response:

Thank you for your suggestions (15–17). We have changed the interpretation of the results from “higher %MoS value indicate good stability” to “a %MoS exceeding 100% suggests instability.” Therefore, the conclusion was changed to “use of the non-paretic leading limb results in lower dynamic stability in the medial direction during the first step. However, there is higher dynamic stability in the lateral direction than with use of the paretic leading limb. Patients with stroke may intentionally maintain such an excessive %MoS to avoid lateral perturbation, which leads to falls.” (lines 33–38, 290–293, 315–322, 346–357, 375–377).

Tables and Figures

1. Figure 3: The standard deviation of the non-paretic limb is hard to distinguish from that of the paretic limb (it’s unclear where the lower bound of the standard deviation of the non-paretic limb is).

Response:

Thank you for your suggestion. We have modified Figure 3 to delete the overlapping shading.

Submitted filename: Response to Reviewers-2022.2.9.docx

Click here for additional data file.

23 Mar 2022

8 Apr 2022

Detailed Response to Reviewers.

Reviewer #3:

We wish to express our deep appreciation for your feedback. As indicated in our point-by-point responses below, we have taken all of your comments and suggestions into consideration in the revised manuscript. Major revisions are highlighted in yellow in the marked-up copy of the manuscript.

Comments

It would be useful to also include results of each subgroup (self-selected paretic leading limb, nonparetic leading limb, mixed leading limb) separately and discuss whether the self-selected leading limb resulted in better balance. At the very least, information on which limb (s) were selected in the three trials with no instruction should be included in the supplementary files. Moreover, the results from each of the three trials for each condition should be included in the supplementary data, not just the average.

This paper is intelligible but it would benefit from additional editing for conciseness and clarity, preferably by a native English speaker.

Response:

Thank you for this feedback. In line with your advice, we performed further statistical calculations on the difference in the leading limb between tasks with and without instruction. However, there was no significant difference between the two. The statistical results are shown in the table below. To avoid making the study overly complex, the additional analysis you suggested has not been added to the main text. The information on which limb was selected in the three trials with or without instruction and raw values before averaging have been added in the supplementary files.

The manuscript has been professionally edited by a native speaker familiar with this area of research. We have attached a certificate of editing.

Leading limb without instruction Leading limb with instruction p-value Effect size (r)

%MoS (%)

%MoS-FO#, median (IQR) 95.1 (22.2) 95.9 (39.6) 0.420 -0.13

%MoS-IC#, median (IQR) 160.0 (28.1) 159.8 (44.3) 0.856 -0.03

#Non-normally distributed; %MoS-FO, %MoS at first foot off; %MoS-IC, %MoS at first initial contact; IQR, interquartile range; MoS, margin of stability.

Line 51 – 54: A citation is needed for this sentence describing the role of the trailing limb, or clarify that the citation is also [7].

Response:

Thank you for noting this. We have clarified the citation in the sentence describing the role of the trailing limb (line 52).

Lines 75 – 76: Citations are needed for “COP movement is the classical method for analyzing movement stability.” To my knowledge, COP movement is used most often for postural stability, not dynamic balance during walking.

Response:

We have changed the sentence from “analyzing movement stability” to “analyzing postural stability” and added citations (lines 75-76).

Lines 75-81: Most of this paragraph justifying the use of MoS over CoP displacement could be removed. MoS is a well-established measurement for dynamic balance and it is clear that MoS is a better measure than CoP movement for this gait initiation task.

Response:

Thank you for this suggestion. We have removed the sentences justifying the use of MoS instead of CoP (lines 76-78).

Line 84: While MoS is a common measurement of dynamic balance, I would refrain from calling it a “gold standard” as there are other measures of balance (such as whole body angular momentum) provide more insight into the underlying causes of balance deficits and are better correlated to clinical test scores, as described in [32]: Vistamehr, A., Kautz, S.A., Bowden, M.G. and Neptune, R.R. (2016). Correlations between measures of dynamic balance in individuals with post-stroke hemiparesis. Journal of Biomechanics 49(3): 369-400.

Response:

We agree with this suggestion and have deleted the phrase “gold standard” (lines 81-82).

Lines 92 – 96: More justification for the hypothesis is needed other than “another study found this”. What are the underlying reasons that the nonparetic leading limb could be more stable?

Response:

To address this comment, we have added the rationale for our hypothesis (lines 89-92).

Line 294 – 332: This is an important but very long paragraph. It should be edited for conciseness and/or divided into shorter paragraphs.

Response:

Thank you for this feedback. We have divided the long paragraph into three parts.

Lines 333 - 345: A discussion of why the pelvic tilt is higher when the paretic leg is the leading limb would be useful. Perhaps hip hiking to lift the paretic leg.

Response:

Thank you for this suggestion. We have added discussion noting that the excessive pelvic tilt is compensatory hip hiking for toe clearance (lines 337-338).

12 Apr 2022

Differences in mediolateral dynamic stability during gait initiation according to whether the non-paretic or paretic leg is used as the leading limb

PONE-D-21-29688R2

Dear Dr. Osada,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Peter Andreas Federolf

Academic Editor

PLOS ONE

18 Apr 2022

PONE-D-21-29688R2

Differences in mediolateral dynamic stability during gait initiation according to whether the non-paretic or paretic leg is used as the leading limb

Dear Dr. Osada:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Peter Andreas Federolf

Academic Editor

PLOS ONE