Split-Belt Treadmill Training Improves Mechanical Energetics and Metabolic Cost in Women with Unilateral Hip Osteoarthritis: A Proof-of-Concept Study

Round 1

Reviewer 1 Report

General comments:

This is an interesting submission which attempted to modify step length symmetry in women with OA to determine whether metabolic cost can further modify gait efficiency. As a proof-of-concept paper, this study demonstrates that the hypothesis is supported by the outcomes and gait efficiency can be modified via step length modification. While I understand this is a proof-of-concept paper, there are still some issues/concerns I have, which I have detailed below. In general, there are several instances where the text is very difficult to read due to poor phrasing. I also encourage the authors to clarify their filtering techniques, as they are briefly described, but may not be appropriate given their instruments.

Introduction:

The sentence in lines 34-35 does not make sense: ‘we have previously demonstrated that greater energy used during walking in women with hip OA was associated with lower self-reported physical activity time…’. Are the authors trying to state that women with hip OA use more energy when walking, and that is due to lower physical activity time? Please clarify.

Rather that stating non-OA or OA leg, it might be more appropriate to use affected or unaffected limb.

In line 45, the authors use the phrase, ‘increased metabolic cost of gait [10-12].’, This might read better if the sentence was edited to ‘increased metabolic gait costs’.

The introduction is well-framed, but there are some semantic issues with phrases and writing that     I suggest editing to make it more readable.

Methods

The authors mention that a 6Hz filter was used on the data, however, I want to clarify if that was for kinematic and kinetic data? If that filter was used on the kinetic data, there is likely a large amount high-frequency noise associated with those signals.

The paragraph describing step length is very wordy and is confusing, even for those who understand how step length is found. I would suggest just stating that step length is found by the heel markers of the contralateral limbs at their respective heel strikes.

In the statistical analysis section, the first sentence (line 151) mentions the use of a paired t-test to compare step length between limbs. I am confused why the authors are testing the differences between step lengths when symmetry index (SI) is also being computed? If SI is the main outcome, do we care about the magnitudes of step length? Also, since step length was measured at four different time periods, this is assuming that data are collapsed to come degree, which is not stated, and is also not recommended. I’d recommend that this analysis be removed since it does not add any scientific merit to the study.

Results

The results are sufficient, however, I am confused why Cohen’s d is being reported? It was not mentioned in the statistical analysis section and readers do not know how those were computed/derived. Please include in the previous section or remove.

Figure 7 y-axis has ‘Change’ misspelled

Discussion

While I agree with the authors that the end of adaptation was different from baseline, I would argue that those findiings are almost moot since post-adaptation values revert back to baseline values. Meaning that yes, you can modify step length symmetry, but the single session does not impose any changes that will remain.

Author Response

Thank you for the thoughtful comments. We have revised the introduction, clarified our methods and statistical analysis, modified figures, and added several limitations to the discussion. Below are the details of responses to the comments.

Introduction:

• The sentence in lines 34-35 does not make sense: ‘we have previously demonstrated that greater energy used during walking in women with hip OA was associated with lower self-reported physical activity time…’. Are the authors trying to state that women with hip OA use more energy when walking, and that is due to lower physical activity time? Please clarify.

In this sentence, we are trying to state that women with hip OA who use more energy when walking were less physically active in a recent cross-sectional study from our group. We have restructured the introduction to more clearly describe the relationships among energy use during walking, gait alterations, and physical activity in women with hip OA. With this additional context, we hope that this statement is clear now. We have modified the paragraphs in line 30-38.

• Rather that stating non-OA or OA leg, it might be more appropriate to use affected or unaffected limb.

We have restructured the introduction and removed that sentence.

• In line 45, the authors use the phrase, ‘increased metabolic cost of gait [10-12].’, This might read better if the sentence was edited to ‘increased metabolic gait costs’.

We have restructured the introduction and removed that sentence.

• The introduction is well-framed, but there are some semantic issues with phrases and writing that I suggest editing to make it more readable.

As suggested, we have modified some issues with phrases and writing in the introduction.

Methods

• The authors mention that a 6Hz filter was used on the data, however, I want to clarify if that was for kinematic and kinetic data? If that filter was used on the kinetic data, there is likely a large amount high-frequency noise associated with those signals.

We used a 6 Hz filter for the kinematic data and a 30 Hz filter for the kinetic data. As suggested, the information has been added to line 107.

• The paragraph describing step length is very wordy and is confusing, even for those who understand how step length is found. I would suggest just stating that step length is found by the heel markers of the contralateral limbs at their respective heel strikes.

       As suggested, the paragraph describing step length has been modified to step length was defined as the distance between the heel markers of each limb at their respective heel strikes in line 111.

• In the statistical analysis section, the first sentence (line 151) mentions the use of a paired t-test to compare step length between limbs. I am confused why the authors are testing the differences between step lengths when symmetry index (SI) is also being computed? If SI is the main outcome, do we care about the magnitudes of step length? Also, since step length was measured at four different time periods, this is assuming that data are collapsed to come degree, which is not stated, and is also not recommended. I’d recommend that this analysis be removed since it does not add any scientific merit to the study.

As suggested, we have removed the use of a paired t-test to compare step length between limbs in the statistical analysis section and the results.

Results

• The results are sufficient, however, I am confused why Cohen’s d is being reported? It was not mentioned in the statistical analysis section and readers do not know how those were computed/derived. Please include in the previous section or remove.

The Cohen’s d was included to give the reader a sense of the magnitude of the change, which is important for this proof-of-concept study. As suggested, we have added the information about Cohen’s d in the statistical analysis section in lines 152-154.

• Figure 7 y-axis has ‘Change’ misspelled

       As suggested, the figure 7 y-axis “Change” has been corrected.

Discussion

• While I agree with the authors that the end of adaptation was different from baseline, I would argue that those findings are almost moot since post-adaptation values revert back to baseline values. Meaning that yes, you can modify step length symmetry, but the single session does not impose any changes that will remain.

We agree with the reviewer and contend that this does not detract from the novelty and potential importance of the study. To our knowledge, this study was the first to attempt split-belt treadmill training to modify step length asymmetry as a strategy to increase gait efficiency in people with hip OA. It was expected that the effects would wash out (de-adaptation) during the early post-adaptation period and return to baseline. Other studies have found that a single exposure to a bout of split-belt treadmill training can have lasting effects for three weeks (Buurke, 2022) and that it takes multiple sessions to see a lasting change that transfers to overground walking (Reisman, 2009). This study introduces a novel strategy to improve gait efficiency in older adults with hip OA to preserve or improve mobility and physical activity levels.

Author Response File: Author Response.pdf

Reviewer 2 Report

The authors conducted a proff-of-concept study to investigate whether altering step length with a split-belt treadmill can improve walking efficiency. Ten women diagnosed with unilateral hip osteoarthritis participated in the study. The intervention was performed in a single session and lasted 15 minutes. Biomechanical and physiological parameters were monitored. The authors concluded that reducing step length asymmetry with split-belt treadmill training can improve walking efficiency in women with hip osteoarthritis.

The authors have done excellent research, for which I must commend them. The introduction is clear and informative, the methods are presented accurately and comprehensively. The results are presented logically and clearly. The figures lack an informative title. The discussion is well written. The only thing missing from the article is mention of hip pain as an important factor that can affect the biomechanics of walking.

Abstract

Line 22: I suggest to the authors to explain what »rho« abbreviation means.

 

Introduction

I suggest to the authors to add meaning of pain in altering gait biomechanics in hip osteoarthritis.

 

Materials and Methods

Table 1:

I suggest to the authors to change the title to Demographic data. Last two lines are not aligned. Since there is no mention that ten participants were recruited (except in text) I suggest putting absolute numbers instead of percentage.

Line 98: Figure 1

I suggest to the authors to leave only the first sentence of the title. The rest of the text is written in the text and there is no need to put it in the Figure title.

 

Line 156: I suggest to the authors to add abbreviation »rho« when first time Spearman correlation coefficient is mentioned.

 

Line 157: Remove one space in front of »An …«.

 

Results

Line 188: Figure 2 need a title. Now only explanation of A and B are written. The explanation of figure B is too much substantively and to less informative about what the figure B represents. The same problem is repeated in all figures.

 

Discussion

I suggest to the authors to mention the possible influence of pain on gait biomechanics in hip osteoarthritis.

Author Response

Reviewer 2

Thank you for the thoughtful comments. We have revised the abstract and introduction, modified the table and figures, and added several limitations to the discussion. Below are the details of responses to the comments.

Abstract

• Line 22: I suggest to the authors to explain what »rho« abbreviation means.

       As suggested, the explanation of rho has been added to the abstract in line 19.

Introduction

• I suggest to the authors to add meaning of pain in altering gait biomechanics in hip osteoarthritis.

In response to this suggestion and one from the other reviewer, we have added context and explanation regarding the association of pain and gait and pain and walking energetics in the Introduction in lines 36-38 and 46-49.

Materials and Methods

• Table 1: I suggest to the authors to change the title to Demographic data. Last two lines are not aligned. Since there is no mention that ten participants were recruited (except in text) I suggest putting absolute numbers instead of percentage.

As suggested, the table 1 title has been modified to Demographic data. The last two lines have been aligned. The percentage also has been modified to absolute numbers.

• Line 98: Figure 1 I suggest to the authors to leave only the first sentence of the title. The rest of the text is written in the text and there is no need to put it in the Figure title.

       As suggested, we only left the first sentence of the title for Figure 1 in line 100.

• Line 156: I suggest to the authors to add abbreviation »rho« when first time Spearman correlation coefficient is mentioned.

As suggested, the abbreviation “rho” has been added to the first time Spearman correlation coefficient is mentioned in line 155.

• Line 157: Remove one space in front of »An …«.

      As suggested, the space in front of “An….” has been removed.

Results

• Line 188: Figure 2 need a title. Now only explanation of A and B are written. The explanation of figure B is too much substantively and to less informative about what the figure B represents. The same problem is repeated in all figures.

      As suggested, we have added the title for Figure 2 and the rest of the figures. We also modified the explanation of the figures.

Discussion

• I suggest to the authors to mention the possible influence of pain on gait biomechanics in hip osteoarthritis.

As suggested, we have mentioned the possible influence of pain on gait biomechanics in hip OA and list it as a limitation for our study in lines 330-334.

 

Author Response File: Author Response.pdf