Antagonist Coactivation of Muscles of Ankle and Thigh in Post-Stroke vs. Healthy Subjects during Sit-to-Stand Task

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

 

The authors present observations on impaired co-contraction of agonist and antagonist muscles during a sit to stand task (SiTS) for persons with subcortical stroke and able-bodied controls. They show that the persons with stroke have significantly higher co-contraction amplitudes relative to controls for both the ipsilesional and contralesional legs. The topic of the manuscript is relevant to journal’s readership and is important to our understanding of impaired neuromuscular recruitment after stroke. 

While the results are interesting, there are problematic methodological concerns that limit the generalizability of the results (see below). Further, the current observations do not provide sufficient mechanistic evidence to support the author’s speculation regarding ipsilateral control dysfunction and the impaired contribution of reticulospinal systems. The manuscript would greatly benefit from a careful grammar/word structure revision to improve readability. The manuscript would also benefit from a greater justification of the methodology to provide a more compelling argument that the co-activation  patterns observed here deviate from healthy coactivation patterns and relevant for the double support phase of gait.

Major concerns outlined below

METHODS/RESULTS

1.      EMG processing concerns:  Important details in the EMG signal analyses are missing. The formula to calculate coactivation is sensitive to EMG normalization process. It is not clear whether the EMG signals have been normalized to the maximum voluntary contraction or another normalization procedure. Fair between subject comparisons of the amplitude responses alone (without normalization) will be skewed given the variability in electrode placement, volitional activation capacity, recording signal noise, etc.…,

2.      It is not clear what type of statistical analyses were performed in the text. It appears that the data analyses structure warrants a repeated measures ANOVA but this is not stated. Further, did the authors utilize a multiple comparisons adjustment and does this affect the significance of the comparisons?

3.      Was the dominant leg of the control group analyzed?

4.      Please label the Figure axis for clarity and increase graph size. Providing a scale legend would improve the interpretation of the figures.

5.      Were both positive and negative deflections in the anterior-posterior force counted toward the onset of the anti-gravity subphase?

6.      At what point in time during each subphase was the coactivation index calculated- and does this choice affect the resulting observations?

7.      It is not clear if normalization for ground reaction forces means that the baseline weight was subtracted from the signal?

8.      The use of upper arms is an important variable that is not controlled here. Do all participants use the same arm strategy for stability?

9.      How did the authors control variability in lower limb impairment within the stroke cohort? I would imagine that even for the same lesion locations and time post stroke, there remains large inherent variability in the muscle strengths across the lower limb as well as walking ability. Further, varying degrees of spasticity may confound the current observations.

 

DISUCSSION/INTERPRETATION

1.      Without comparison to other functional metrics (walking speed, leg strength), it is difficult to know if SiTS coactivation results will generalize to  waling ‘incapacity’

2.      Mechanisms underlying IPSI vs CONTRA weakness merit further discussion, especially if the authors are suggesting that the IPS presents the most relevant alteration.

3.      It is unclear what the ‘variation’ of expression of coactivation indices means.

4.      It is not clear that the coactivation indices differ between the subphases in the stroke cohort. Thus, the statement that ankle muscle plays a determinant role of the proximal muscle activity is hard to follow if based on current observations

Comments on the Quality of English Language

There are grammar issues throughout the manuscript that should be revised.

Author Response

We are grateful to the Reviewer for his/her comments and suggestions that have helped us improving our work. We have taken into account the comments made. Below we provide more detailed explanations of these changes.

Please note that the changes that we have made are highlighted in the revised version (througth “track changes” function).

 

Comments and Suggestions for Authors

 

The authors present observations on impaired co-contraction of agonist and antagonist muscles during a sit to stand task (SiTS) for persons with subcortical stroke and able-bodied controls. They show that the persons with stroke have significantly higher co-contraction amplitudes relative to controls for both the ipsilesional and contralesional legs. The topic of the manuscript is relevant to journal’s readership and is important to our understanding of impaired neuromuscular recruitment after stroke. 

While the results are interesting, there are problematic methodological concerns that limit the generalizability of the results (see below). Further, the current observations do not provide sufficient mechanistic evidence to support the author’s speculation regarding ipsilateral control dysfunction and the impaired contribution of reticulospinal systems. The manuscript would greatly benefit from a careful grammar/word structure revision to improve readability. The manuscript would also benefit from a greater justification of the methodology to provide a more compelling argument that the co-activation  patterns observed here deviate from healthy coactivation patterns and relevant for the double support phase of gait.

 

Authors: We would like to thank the Reviewer the positive comment about the manuscript.

Bellow we present the responses to the issues raised by the Reviewer.

 

Major concerns outlined below.

Methods/Results

Point 1: EMG processing concerns:  Important details in the EMG signal analyses are missing. The formula to calculate coactivation is sensitive to EMG normalization process. It is not clear whether the EMG signals have been normalized to the maximum voluntary contraction or another normalization procedure. Fair between subject comparisons of the amplitude responses alone (without normalization) will be skewed given the variability in electrode placement, volitional activation capacity, recording signal noise, etc.…

Response 1: We have revised this point, and the important details in the EMG signal analyses can be find in manuscript, in line 123. The EMG signals have been normalized to the maximum voluntary contraction of the subjects’ different muscles.

 

Point 2: It is not clear what type of statistical analyses were performed in the text. It appears that the data analyses structure warrants a repeated measures ANOVA but this is not stated. Further, did the authors utilize a multiple comparisons adjustment and does this affect the significance of the comparisons?

Response 2: We have revised this point and better clarified the statistical analyses. It can be find in the manuscript, in line 165. The authors, to ensure that there were no significant differences between groups (stroke vs. healthy) regarding age, height and weight, used the independent T-Test. Since a normal distribution was not verified on all the coactivation variables, Kruskal Wallis Test was applied to compare antagonist coactivation levels between Healthy, CONTRA and IPSI limbs. The Mann-Whitney Test was used to compare antagonist coactivation levels between Healthy and IPSI, Healthy and CONTRA, and IPSI and CONTRA limbs. A significance of 0.05 was considered for analysis.

 

Point 3: Was the dominant leg of the control group analyzed?

Response 3: In the control group, just one lower limb was randomly selected to be analysed. This information can be stated in the manuscript in line 120.

 

Point 4: Please label the Figure axis for clarity and increase graph size. Providing a scale legend would improve the interpretation of the figures.

Response 4: We appreciate the suggestion. Figure 1 was improved regarding labels and size (line 129). Since this figure is a representative ilustration of the followed strategy to define the sub-phases A and B, and is not a real output, we think it is not necessary to provide a scale.

 

Point 5: Were both positive and negative deflections in the anterior-posterior force counted toward the onset of the anti-gravity subphase?

Response 5: We thank the opportunity to clarify the observation. Yes, both deflections were counted. This procedure is described from line 128 to 142.

 

Point 6: At what point in time during each subphase was the coactivation index calculated- and does this choice affect the resulting observations?

Response 6: In each sub-phase the mean of the EMG signal was used to assess antagonist coactivation through the coactivation formula (line 148).

 

Point 7: It is not clear if normalization for ground reaction forces means that the baseline weight was subtracted from the signal?

Response 7: Yes, the values were normalized to the weight of each subject, by subtracting their weight to the baseline (line 125).

 

Point 8: The use of upper arms is an important variable that is not controlled here. Do all participants use the same arm strategy for stability?

Response 8: Knowing that the use of the upper arms is an important variable to be concerned about,

subjects were asked to stand up without using them (stated in the manuscript in line 116), maintaining them confortably along the body/trunk.

 

Point 9: How did the authors control variability in lower limb impairment within the stroke cohort? I would imagine that even for the same lesion locations and time post stroke, there remains large inherent variability in the muscle strengths across the lower limb as well as walking ability. Further, varying degrees of spasticity may confound the current observations.

Response 9: Thank you for the appreciation. The Fugl-Meyer Assessment of Sensorimotor Recovery After Stroke Scale was applied in the post-stroke subjects and, to minimize lower limb variability impairment, they were included if they have score below 34 in the motor subsection (Duncan, Propst, & Nelson, 1983, Lamontagne et al. 2002). This information was added in line 84 of the manuscript.

 

Discussion/Interpretation

Point 1: Without comparison to other functional metrics (walking speed, leg strength), it is difficult to know if SiTS coactivation results will generalize to walking ‘incapacity’

Response 1: We thank the observation. It was better clarified in the manuscript, in line 214.

Despite the absence of other functional variables related to gait, given the scarcity of specific information on the coactivation analysis studied in the sit-to-stand task, and knowing that, as in the double-support phase of walking, where both sides seems to reflect a bilateral postural control dysfunction in these subjects, and there’s a precise control of joint position, influenced by supra-spinal structures on antagonist coactivation that reinforces the value of this synergy in joint stability, seems that the comparision may be adequate. The findings obtained agree with previous studies involving the antagonist coactivation walking, although future studies, with more metrics, are needed to this assumption.

 

Point 2: Mechanisms underlying IPSI vs CONTRA weakness merit further discussion, especially if the authors are suggesting that the IPS presents the most relevant alteration.

Response 2: We thank the important observation. We took it into consideration (line 233) regarding that although these assumptions, it is important to think about the mechanisms underlying IPSI vs CONTRA weakness, while suggesting that the IPSI presents the most relevant alteration in antagonist coactivation.

 

Point 3: It is unclear what the ‘variation’ of expression of coactivation indices means.

Response 3: We thank the observation. The setence was not properly build. ment to say “variations expressed on coactivations”. It was corrected in the manuscript, in line 253.

 

Point 4: It is not clear that the coactivation indices differ between the subphases in the stroke cohort. Thus, the statement that ankle muscle plays a determinant role of the proximal muscle activity is hard to follow if based on current observations

Response 4: We thank the observation, and we tried to improve it according to the current findings, in the manuscript, line 244.

It was observed that there were no significant differences in the antigravity sub-phase, in the antagonist coactivation of the thigh, while in leg it was stated.

Knowing that ankle muscles, specifically the plantar flexors, have an important role improving the ankle postural control performance strategies to get standing, it can be suggested, according to previous studies (Yang, et al, 1990, Fitzpatrick, Douglas, & McCloskey, 1994; Loram, Maganaris, & Lakie, 2005), why the differences were mostly observed distally. Ankle coactivation role for stabilization during movement in space may have a major influence on the proximal adjustment, explaining the differences, expressing a higher performing postural control demand. The non-existence of proximal statistical siginificative differences may demonstrate the lower requirement of postural performance during the task, when compared to distal, observed by the results.

 

Comments on the Quality of English Language

There are grammar issues throughout the manuscript that should be revised.

Response: We would like to thank the Reviewer for the comment about the grammar issues. The manuscript was all carefully revised for english revisions.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

This study examines the level of co-activation of leg muscles during sit-to-stand motions in stroke patients. The reviewers felt that this trial would be very beneficial clinically. This study includes many stroke patients. On the other hand, the calculation of co-activation levels is classic. Whether this simple formula was valid depends on the actual time series of muscle activity. The authors should be more forthcoming with their data. This will lead to a better understanding by the readers.

 

Below are comments for each specific point.

 

[Methods]

L108 Reducing skin resistance is important in detecting muscle activity. However, reviewers know that it is often difficult in patients. Indeed, what procedures did the authors use to reduce skin resistance in patients? This information will be useful for readers to apply this knowledge to their clinical practice.

 

L115 The authors stated that "sufficient repetitions were performed to obtain three valid trials." How many trials actually failed? And the way patients stand up is diverse; L121 classifies movements based on floor reaction force, but the reviewers wonder if it worked in all trials. The authors should describe in more detail how they defined success/failure.

 

L133 The formula for calculating the co-activation level employs the average of muscle activity during a given phase. The muscle may have been steeply active at some point during the phase. Or the muscle might have averaged activity throughout the phase. In both cases, the results converge to the same average. The reviewer wonders if the co-activation level truly reflects simultaneous contractions between the muscles. For example, if muscle A has steep activity at one time during a phase and muscle B has steep activity at another time, muscle A-B would be quantified as co-activated. The authors should present a time series of actual muscle activity in healthy subjects and stroke patients, at least one case each, to prove that this was not the case. Or are there such figures in the Supplementary Materials of L239? The reviewer could not read those figures.

 

L150 Why did the authors use non-parametric tests? If a test of normality was used, please mention the results.

 

[Discussion]

L199 and L209 The authors considered muscle co-activation as linked to impairment of specific brain regions. However, the present study did not examine this as such, and the previous study cited by the authors is a different task. It is theoretically acceptable for the authors to mention a specific brain region as a hypothesis, but it should be stated as a LIMITATION of the study because we do not know if it is true.

 

L199 In addition, it would be a leap in logic to consider whether muscle co-activation is truly acting as a postural stabilizer in the sit-to-stand motion from a different motion. When adding such a consideration, it is appropriate to cite simulation studies of muscle synergy in the sit-to-stand motion. The reviewers felt that Qi An's series of studies fit this consideration (e.g., An, Qi, et al. 2013 IEEE Workshop on Advanced Robotics and its Social Impacts. IEEE, 2013.)

Author Response

We are grateful to the Reviewer for his/her comments and suggestions that have helped us improving our work. We have taken into account the comments made. Below we provide more detailed explanations of these changes.

Please note that the changes that we have made are highlighted in the revised version (througth “track changes” function).

 

Comments and Suggestions for Authors

 

This study examines the level of co-activation of leg muscles during sit-to-stand motions in stroke patients. The reviewers felt that this trial would be very beneficial clinically. This study includes many stroke patients. On the other hand, the calculation of co-activation levels is classic. Whether this simple formula was valid depends on the actual time series of muscle activity. The authors should be more forthcoming with their data. This will lead to a better understanding by the readers.

 

Authors: We would like to thank the Reviewer the positive comment about the manuscript.

Bellow we present the responses to the issues raised by the Reviewer.

 

Below are comments for each specific point.

Methods

Point 1: L108 Reducing skin resistance is important in detecting muscle activity. However, reviewers know that it is often difficult in patients. Indeed, what procedures did the authors use to reduce skin resistance in patients? This information will be useful for readers to apply this knowledge to their clinical practice.

Response 1: Thank you for the observation. We have revised this point, and the proper procedures were better described in the manuscript, in line 109.

The skin surface of selected muscles midbelly and the local of ground electrode was prepared through standards procedures (shaving, removing dead skin cells and non-conductor elements with alcohol and with an abrasive pad), to reduce the electrical resistance to <5 KΩ (Correia, 2004; Sousa, Silva, Santos, et al., 2013; Sousa, Silva, & Tavares, 2013), monitored with an Electrode Impedance Checker®, Noraxon® device (Noraxon, Scottsdale Arizona, USA).

 

Point 2: L115 The authors stated that "sufficient repetitions were performed to obtain three valid trials." How many trials actually failed? And the way patients stand up is diverse; L121 classifies movements based on floor reaction force, but the reviewers wonder if it worked in all trials. The authors should describe in more detail how they defined success/failure.

Response 2: We thank the question. We try to justify it in line 116. A successful trial was considered when the subject completed the requisites definied as no use of the upper limbs, while maintaining them comfortably along the body, no moving the feet, and remaining quietly standing at the end of the task. No more than 5 trials were performed for each subject.

 

Point 3: L133 The formula for calculating the co-activation level employs the average of muscle activity during a given phase. The muscle may have been steeply active at some point during the phase. Or the muscle might have averaged activity throughout the phase. In both cases, the results converge to the same average. The reviewer wonders if the co-activation level truly reflects simultaneous contractions between the muscles. For example, if muscle A has steep activity at one time during a phase and muscle B has steep activity at another time, muscle A-B would be quantified as co-activated. The authors should present a time series of actual muscle activity in healthy subjects and stroke patients, at least one case each, to prove that this was not the case. Or are there such figures in the Supplementary Materials of L239? The reviewer could not read those figures.

Response 3: As suggested by the Reviewer, one example of each case was presented in figure 2 (line 140), demonstranting a time series of muscle activity.

 

Point 4: L150 Why did the authors use non-parametric tests? If a test of normality was used, please mention the results.

Response 4: Statistics topic in the manuscript was improved to better clarify this question (line 160). Non-parametric tests were used because it was not verified a normal distribution on all the coactivation variables, considering a significance of 0.05 for analysis, using the Shapiro-Wilk Test.

 

Discussion

Point 1: L199 and L209 The authors considered muscle co-activation as linked to impairment of specific brain regions. However, the present study did not examine this as such, and the previous study cited by the authors is a different task. It is theoretically acceptable for the authors to mention a specific brain region as a hypothesis, but it should be stated as a LIMITATION of the study because we do not know if it is true.

Response 1: We thank the observation, that was taking into account. This aspect was mentioned in line 253.

 

Point 2: L199 In addition, it would be a leap in logic to consider whether muscle co-activation is truly acting as a postural stabilizer in the sit-to-stand motion from a different motion. When adding such a consideration, it is appropriate to cite simulation studies of muscle synergy in the sit-to-stand motion. The reviewers felt that Qi An's series of studies fit this consideration (e.g., An, Qi, et al. 2013 IEEE Workshop on Advanced Robotics and its Social Impacts. IEEE, 2013.)

Response 2: We appreciate the suggestion. The citation was added to the manuscript in line 215.

 

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

 

The author have addressed many of the concerns in the original submission and have significantly improved the manuscript.

However, the following remaining issues should be addressed to further improve the clarity of presentation.

-Using the term ‘leg level’ in contrast with thigh segment may be hard to interpret by other readers, as both are at the leg. Consider being more specific (e.g., proximal vs distal / ankle joint) to help improve clarity throughout the manuscript.

- I would split lines 114 to 117 into 2 sentences to improve clarity.

- It is not clear from the revision comments if and how the authors adjusted the p values for multiple comparisons to control for Type 1 error rate. If this is not used, please justify and/or state in text.

-Please state that Figure 1 is an illustration/schematic and not real data if this is the case. Otherwise, a vertical scale with units is necessary.

-Figures 1 & 2 look blurry. Figure 2 x axis is not readable.  I strongly suggest further improving the image quality and font size of each figure.

-For clarity, I would split lines 211 to 215 into separate sentences.

-Please revise 233:  “ Although IPSI presents the most relevant alteration in  antagonist coactivation,  future study will….”

-Sentence in line 241- 245 should be revised for clarity – possibly into 2 separate sentences. What does ‘in space’ refer to?

-Line 252:  “Variations observed in coactivation may also improve”

-Improve clarity of line 254: “Considering that all post-stroke subjects analyzed presented with a lesion…the association between”

Comments on the Quality of English Language

Many sentences should be split to improve readability. Suggest reviewing revision 2 additions for quality.

Author Response

We are grateful again to the Reviewer for his/her comments and suggestions that have helped us improving our work. We have taken into account the comments made. Below we provide more detailed explanations of these changes.

Please note that the changes that we have made are highlighted in the revised version (througth “track changes” function).

 

Comments and Suggestions for Authors

 

The author have addressed many of the concerns in the original submission and have significantly improved the manuscript.

However, the following remaining issues should be addressed to further improve the clarity of presentation.

 

Authors: We would like to thank the Reviewer the positive comment about the improved manuscript.

Bellow we present the responses to the issues still raised by the Reviewer.

 

Major concerns outlined below.

Methods/Results

Point 1: Using the term ‘leg level’ in contrast with thigh segment may be hard to interpret by other readers, as both are at the leg. Consider being more specific (e.g., proximal vs distal / ankle joint) to help improve clarity throughout the manuscript.

Response 1: We thank the suggestion that was taken into account. All manuscript was revised to improve clarity. Changes ares highlighted and identified by “Reviewer 1. Point 1” along the manuscript.

 

Point 2: I would split lines 114 to 117 into 2 sentences to improve clarity

Response 2: The sentence was splitted as suggested, to improve clarity (lines 114 to 118).

 

Point 3: It is not clear from the revision comments if and how the authors adjusted the p values for multiple comparisons to control for Type 1 error rate. If this is not used, please justify and/or state in text.

Response 3:

We thank the Reviewer for the observation, and we will try to improve clarity about the comment.

To control Type I error rate and considering that a normal distribution was not verified on all the coactivation variables, it was used the Dunn-Bonferroni post hoc test. This information was stated in the manuscript in line 168.

 

Point 4: Please state that Figure 1 is an illustration/schematic and not real data if this is the case. Otherwise, a vertical scale with units is necessary.

Response 4: As suggested, it was stated that Figure 1 is an illustration (lines 129 and 134).

 

Point 5: Figures 1 & 2 look blurry. Figure 2 x axis is not readable.  I strongly suggest further improving the image quality and font size of each figure.

Response 5: We appreciate the Reviewer’s suggestion and tooke it into account. Figure 1 and 3 had image quality improving, and Figure 2 was replaced.

 

Point 6: For clarity, I would split lines 211 to 215 into separate sentences.

Response 6: As suggested, to improve clarity, lines 210-214 were splited.

 

Point 7: Please revise 233:  “ Although IPSI presents the most relevant alteration in  antagonist coactivation,  future study will….”

Response 7: As the Reviewer suggested, the sentence was revised (line 233).

 

Point 8: Sentence in line 241- 245 should be revised for clarity – possibly into 2 separate sentences. What does ‘in space’ refer to?

Response 8: The sentence was revised and separated as suggested by the Revisor (line 242).

“In space” was refering to the 3D plans where the task was being carried out. But, the sentence was corrected without it, to improve clarity (line 241).

 

Point 9: Line 252: “Variations observed in coactivation may also improve”

Response 9: The sentence was improved to be better understood (line250).

 

Point 10: Improve clarity of line 254: “Considering that all post-stroke subjects analyzed presented with a lesion…the association between”

Response 10: As suggested by the Reviewer, clarity in this line was improved (line 254).

 

 

Comments on the Quality of English Language

Many sentences should be split to improve readability. Suggest reviewing revision 2 additions for quality.

Response: We thank the observation. All the reviewing suggestions were taking into account, in order to provide readability quality.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

The reviewers respect the authors' efforts to improve the validity of their findings. The study is now more methodologically valid. The presentation of actual results from healthy adults and stroke patients makes it easier for readers to understand the results. It also makes it easier for the reader to apply the results to clinical practice. However, one doubt remains in the interpretation of the results, which the authors need to confirm.

On the other hand, the authors should be aware that this study only measured muscle activity and floor reaction forces. Therefore, the authors should be extra cautious in referring to the mechanical role and/or physiological origin of muscle activity. As a result of our comments in the previous peer review, leaps in logic have been reduced, but they still remain.

 

Please refer to the point-by-point comments below.

 

Methods

Figure 2.

Is it really a healthy adult on the left side and a stroke patient on the right side? The reviewers are doubtful. Referring to the representative studies cited by the authors [4, 26, 27], TA activity preceded SOL (as the authors themselves state in L156). However, on the left side of Figure 2 this order is opposite. Also, there is usually not much sustained TA activity in stationary standing. It is understandable that there is alternating TA-SOL activity for anterior-posterior sway immediately after standing up. However, the time range in the figure presented by the author was 6 seconds, which was sufficient for stabilization after standing up. The left side of Figure 2 shows TA activity for a much longer time, which is unnatural.

　More critically, referring only to this figure, it appears that the left side has greater TA-SOL co-activity in phase A. The right side clearly has greater TA activity and less SOL activity in phase A until the floor reaction force onset. This is the opposite of the population trend described by the authors.

　Assuming that the left side is really healthy adults and the right side is stroke patients, it is not good for the paper to present a figure that confuses the interpretation of the results. Please present a figure that is more in line with the population results.

 

Discussion

L216

　There is still a leap in logic that the reviewer commented on in the first review. The Qi An muscle synergy simulation presented by the author is for a healthy adult and does not mention the co-activity of ankle muscles. Rather, the simulations demonstrate that the TA in phase A would be solely responsible for ankle joint stability. Qi-An's series of studies have also been conducted on stroke patients, where the co-activation of ankle muscles has been revealed by muscle synergy studies. Please cite the content of previous studies in appropriate situations.

　Also, it is inappropriate to generalize this to the sit-to-stand motion, since the authors are still talking "only" about walking studies and more basic studies at the joint motion level. The same is true for L229. There are other studies that present the co-activation of ankle muscles in the sit-to-stand motions of stroke patients, as mentioned above. Therefore, the author should actively refer to those studies.

 

L230

　The authors make the same error from a physiological aspect as the mechanical perspective described above. The neurological contributions to SOL cited here are only validated by walking and more basic studies at the joint movement level. The authors cannot generalize to sit-to-stand motions based on those studies alone. There are some prior studies that have validated this for the sit-to-stand motion. Citation 26, "Goulart F, et al. 1999" is a study that confirmed the responsiveness of TA and SOL to transcranial/peripheral nerve electrical stimulation. They examine the cortical/brainstem (essentially at the spinal level) contribution to the ankle muscles in the sit-to-stand motion, as the author would like to describe. This is a prior study on the sit-to-stand motion. We believe that this reinforces the authors' argument more directly than studies on walking movements.

　And the authors should be aware that it will help the reader's understanding if they do not simply cite the literature number, but also outline what is in it.

Comments for author File: Comments.pdf

Author Response

We are grateful again to the Reviewer for his/her comments and suggestions that have helped us improving our work. We have taken into account the comments made. Below we provide more detailed explanations of these changes.

Please note that the changes that we have made are highlighted in the revised version (througth “track changes” function).

 

Comments and Suggestions for Authors

 

The reviewers respect the authors' efforts to improve the validity of their findings. The study is now more methodologically valid. The presentation of actual results from healthy adults and stroke patients makes it easier for readers to understand the results. It also makes it easier for the reader to apply the results to clinical practice. However, one doubt remains in the interpretation of the results, which the authors need to confirm.

On the other hand, the authors should be aware that this study only measured muscle activity and floor reaction forces. Therefore, the authors should be extra cautious in referring to the mechanical role and/or physiological origin of muscle activity. As a result of our comments in the previous peer review, leaps in logic have been reduced, but they still remain.

 

Authors: We would like to thank the Reviewer the positive comment about the improved manuscript.

Bellow we present the responses to the issues still raised by the Reviewer.

 

Please refer to the point-by-point comments below.

Methods

Figure 2

Point 1: Is it really a healthy adult on the left side and a stroke patient on the right side? The reviewers are doubtful. Referring to the representative studies cited by the authors [4, 26, 27], TA activity preceded SOL (as the authors themselves state in L156). However, on the left side of Figure 2 this order is opposite. Also, there is usually not much sustained TA activity in stationary standing. It is understandable that there is alternating TA-SOL activity for anterior-posterior sway immediately after standing up. However, the time range in the figure presented by the author was 6 seconds, which was sufficient for stabilization after standing up. The left side of Figure 2 shows TA activity for a much longer time, which is unnatural.

More critically, referring only to this figure, it appears that the left side has greater TA-SOL co-activity in phase A. The right side clearly has greater TA activity and less SOL activity in phase A until the floor reaction force onset. This is the opposite of the population trend described by the authors.

Assuming that the left side is really healthy adults and the right side is stroke patients, it is not good for the paper to present a figure that confuses the interpretation of the results. Please present a figure that is more in line with the population results.

Response 1: We thank the opportunity to correct the information. The Reviewer comment is completely right. Actually, we have mistaken signals when building the two graphs. In the new graphs version (line 142), we take just 2 seconds of signal (a more specific timeframe from where we obtained data to calculate antagonist co-activation) with EMG signals already in RMS format. As can be seen, the signals from the healthy subject sample are more accordant with the literature (although variability exists among subjects), and stroke signals sample from the contralesional subject´s limb clearly represent the co-activation data obtained (although again, variability among subjects exists).

 

Discussion

L216

Point 2: There is still a leap in logic that the reviewer commented on in the first review. The Qi An muscle synergy simulation presented by the author is for a healthy adult and does not mention the co-activity of ankle muscles. Rather, the simulations demonstrate that the TA in phase A would be solely responsible for ankle joint stability. Qi-An's series of studies have also been conducted on stroke patients, where the co-activation of ankle muscles has been revealed by muscle synergy studies. Please cite the content of previous studies in appropriate situations.

Response 2: We thank the Reviewer’s observation. We improved the manuscript with other studies conducted by Qi-An, given its relevance and value. The citations were stated in the appropriate situations (line 212-229).

 

Point 3: Also, it is inappropriate to generalize this to the sit-to-stand motion, since the authors are still talking "only" about walking studies and more basic studies at the joint motion level. The same is true for L229. There are other studies that present the co-activation of ankle muscles in the sit-to-stand motions of stroke patients, as mentioned above. Therefore, the author should actively refer to those studies.

Response 3: As suggested, other studies in sit to stand with stroke were referenced to improve the statement (line 225).

 

L230

Point 4: The authors make the same error from a physiological aspect as the mechanical perspective described above. The neurological contributions to SOL cited here are only validated by walking and more basic studies at the joint movement level. The authors cannot generalize to sit-to-stand motions based on those studies alone. There are some prior studies that have validated this for the sit-to-stand motion. Citation 26, "Goulart F, et al. 1999" is a study that confirmed the responsiveness of TA and SOL to transcranial/peripheral nerve electrical stimulation. They examine the cortical/brainstem (essentially at the spinal level) contribution to the ankle muscles in the sit-to-stand motion, as the author would like to describe. This is a prior study on the sit-to-stand motion. We believe that this reinforces the authors' argument more directly than studies on walking movements.

Response 4: We thank the Reviewer for the observation. We have improved the description made in the manuscript with the proper citation (lines 228 and 230).

 

Point 5: And the authors should be aware that it will help the reader's understanding if they do not simply cite the literature number, but also outline what is in it.

Response 5: We thank the observation and tried to improve the manuscript according to it.

Author Response File: Author Response.pdf

Round 3

Reviewer 1 Report

Comments and Suggestions for Authors

The authors have sufficiently addressed all remaining comments and concerns. No further revisions are suggested.

Reviewer 2 Report

Comments and Suggestions for Authors

The authors answered all of our concerns.
The reviewers see no problems with the quality of this manuscript.