A Comparison of Motor Unit Control Strategies between Two Different Isometric Tasks

Round 1

Reviewer 1 Report

Overall, this study is well presented and clear. New findings are proposed.

Introduction

The rationale of the study is clearly exposed mainly for one independent factor: isometric conditions (force task vs. position task). One first thing needs to be considered when authors are presenting the scientific background of their study: please comment whether the results presented in the scientific background were found on female, male or both; any potential differences? The second point is on the second factor authors are manipulating: contraction intensity (%MVIC). The rationale is lacking and goal/hypothesis are not focus on this factor (and the interaction factor). Please revise accordingly.

Methods

L96. Give information on how the laterality was determined.

L100. Men and women were recruited. Does the literature clearly indicate no effect of this criterion on the main variables of interest from surface EMG signals?

The description of the protocol is clear.

L164. How was reproducible after the familiarization phase the force values at maximal and submaximal levels? Can you indicate the coefficient of variation for instance for a couple of force measures? The setup as visible in Figure 1 shows that there are many degree of freedom of movement for producing the force tasks. In this case, have you controlled (measured) other muscle mass or group with EMG data collection?

L170-175. A figure or picture showing the locations of EMG electrodes and their montage over elbow flexors and extensors could be proposed for the readers.

L178. Did you measure impedance values for each EMG electrodes? and have you used a target value as it is usually recommended?

L211. What does mean “approximately”? Normality should be ok or not. Please clarify if some distributions were not following a normal law.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Manuscript Number:  ijerph-746419

Title: A comparison of motor unit control strategies between two different isometric tasks

GENERAL COMMENTS

The manuscript is about MU control strategies for non-fatiguing isometric elbow exercises at 40% and 70% MVIC.

Although the topic is not very relevant to the readers, the paper is well organized and written.

I have some major comments:

1. In my opinion, the most critical issue of the study is the standardization of the procedure to test the elbow flexion (position task) between the subjects. I wonder how the participants could hold the elbow angle position of 135° without the forearm fixed, throughout the task. Is the contraction really isometric? This issue has also effects on the results of the study.
2. In the Abstract and in the Materials and Methods sections it is stated that 19 subjects participated in the study, even though in Table 1 and Table 2 results for only 15 subjects are presented. What happened to the other 4?
3. The description of the experimental procedures is very complicated (starting from P3 L123). I would suggest the authors to add at least two pictures, similar to those provided in their recent publication in Physiological Reports. Moreover, since this journal is not used to publish many EMG/decomposition studies, I kindly suggest the authors to provide a picture with the electrode position on the selected muscles.
4. In the end of the conclusions, I suggest the authors to add few lines about future studies to determine the mechanism for the different MU control strategies.

Minor comments

Throughout the manuscript motor unit may be abbreviated to MU.

P2 L60-61 This sentence is not clear. Please rephrase

P3 L104-108 The sentence in brackets is rather long. Please reword.

P10 L342-347: Not clear, please rephrase.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

Figure 1: The Delsys sensors are not shown in the figure 1. The location of the sensor is not defined in the text. Where is the sensor location on the surface of the muscle? On the muscle belly? Margin or mid-line or close to the muscle-tendon junction? Was the location kept persistent between different EMG recordings?

Figure 4. It is very difficult to understand why de-recruitment threshold can go to negative reading.  Personally, there is always baseline activity of muscle firings, it is impossible to be less than 0. 

Table 1. It is  not clear how the number of MU was obtained? It is very difficult to accept that only 20 ± 7 of MUs were recruited during 40% of MVIC contraction.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 4 Report

This study assessed motor units data from the biceps brachii muscle of 19 subjects during two isometric tasks; positional and force trials.  Mean firing rate, recruitment threshold and derecruitment thresholds were assessed for motor units decomposed from surface EMGs collected with a 5-pin detection system.  Key results revealed the relationship between motor units’ data was not comparable between force and position trials.  This led the authors to conclude that, for a given force level, position tasks may depend more heavily on the recruitment of high-threshold units than force tasks.  Even though I find these results original, I have general and specific major, conceptual and methodological concerns.

Section 1 poses more a state of the art description than Introduction.  Instead of focussing on describing results from previous studies, the authors should provide readers with statements clearly highlighting: i) a relevant, open issue; ii) the specific question the authors wish to address and iii) a motivated hypothesis.  Stating “it may also be interesting to examine …” or that “the direct comparison of motor units … has not been examined” does not motivate the study.  Accordingly, “to examine the relationship between the motor unit RT and MFR” is a rather vague objective.  What is the specific research question the authors wish to address here?  How is addressing this question expected to advance our knowledge with respect to the state of the art?  What is the most expected answer to this question and why?  Provide readers with elements indicating this study does indeed contribute to advancing our current knowledge on a specific issue.  When doing so, the authors should clearly indicate the novelty of their work.  The effect of force level and limb posture on differences in control strategies between positional and force tasks has been well discussed in the review by Enoka et al [1] (cf. section 3.1).  Even though this discussion was mostly focused on fatiguing contractions, the authors must indicate readers what is the new, specific knowledge their manuscript is expected to provide.

Another major point regards how comparable data from force and position tasks are comparable.  Although I understand the authors have used a load cell to control for the force values between the two tasks, they did not show how much the rate of increase in force is comparable for the two tasks.  During the position task force was applied by an experimenter while the subject attempted to maintain the elbow joint at a fixed angle.  No objective data has been shown however showing position was indeed kept constant throughout the whole trapezoidal trial and that force increased similarly in both force and position tasks.  Without showing that subjects produced similar force profiles during both trials, while keeping variations in elbow joint position to a minimally acceptable value (130-140°?), no inferences on differences in control strategies between tasks may be advanced from the current data.

Specific issues

Section 1

Statement in lines 29-31 is awkward.  Isometric muscle contraction is not responsible for performing any task.  Isometric contraction implies no change in muscle length, regardless of the circumstances.

Lines 35-36: Is this a deduction or are the authors able to name and motivate these factors?

Line 58-59: Is verb tense set correctly?

Lines 60-61: A clause seems to be missing here.

Line 61: Low-threshold motor units have been shown to achieve and not have higher firing rates.

What is “the neural hand position”? (Line 129)

“supporting the wrist of the dominant wrist.” is an awckward construction (Line 130)

Be consistent when reporting values “3-sec” (Line 132) “5-second” (Line 137).

Why different durations were applied for measuring MVCs for elbow extensors and flexors?

Section 2.3

How was EMG amplitude computed during the MVC tasks?  I see a 1s epoch was used but it is not clear how this window has been moved over the entire 3s contractions for triceps.  Were for example three RMS values computed over the 3s trial and the highest value retained or did the authors move the 1s window assuming a given overlap between them in search for the highest RMS value?  Why RMS values were computed only for the triceps brachii during MVCs?  This is in contradiction with the normalisation procedure commented in Lines 196-197.

How did the authors ensure the rate of increase in force was comparable in both force and position tasks?

How can the authors ensure variations in joint angle were due to the subject being tested and not to the experimenter holding the handle that, in turn, controlled the position of the joint angle?

Section 2.4

If bipolar electrodes were placed on the belly of biceps brachii, where did the authors position the 5-pin detection system used for decomposition?

What was the dynamic range of the A/D converter used to sample EMGs?  Considering the 1,000 gain and the high contraction levels applied in this study, the authors must explicitly indicate no saturation was observed in any of the collected signals.  Showing raw data for any of the high-contraction level tasks could help.

The sentence “… the firing rate curve of each motor unit was plotted with a 1-s Hanning window” is unclear.  What is meant by this?  Discharge instants take place at irregular time intervals and therefore the temporal distance between consecutive discharge rate values is not constant.  How did the authors produce the smooth discharge rate profiles shown in Figure 2?

In Line 189 you mean “in relation to the percentage of the maximal isometric strength?”

How was RT and DT specifically defined?  How many times did the subjects repeat the trapezoidal contractions, both for the force and position tasks?  Given the variable number of motor units identified and the variable coefficient of determination that the authors appear to have obtained, it is unclear how reproducible the RT and DT values may be within subjects.

Red and light blue traces shown in Figure 2 are unclear.  How do the authors explain and increase in discharge rate from roughly 2 to 8 pps without a corresponding increase in muscle force?  Considering these are likely large units, given they were the last two recruited units, would not such an increase in discharge rate results in an appreciable increase in force?  Did the authors inspect the spike-triggered average action potentials for these two units?  Also, 2 pps seems a rather unphysiological value; motor unit do not appear to discharge regularly below 4 pps [2].

Section 3.2

Why 17 and then 15 subjects?  When did the authors report to have discarded subjects?  Why were subjects discarded?

“when compared” (Line 248)

Slope values must be indicated with their unit of measurement when referring to MFR and RT analyses.

Figures 3 and 4 must be improved for readability.  Characters are barely legible.  Also, three plots in Figure 4 are titled with 40% MVIC labels

What is the physical meaning of negative values for DT in Figure 4?  How can a motor units be derecruited at negative force levels?

Section 4

Statement in line 321-322 must be softened.  Differences in EMG amplitude between contraction levels have been recently reported for biceps brachii motor units [3].  Moreover, the 5-pin and the bipolar electrodes were not positioned at the same muscle region and therefore no statements can be advanced on the comparability of information extracted by both detection systems.  EMGs detected from different muscle regions do not necessarily sample from the same muscle fibres [3,4].

References

[1]      R.M. Enoka, S. Baudry, T. Rudroff, D. Farina, M. Klass, J. Duchateau, Unraveling the neurophysiology of muscle fatigue, J. Electromyogr. Kinesiol. 21 (2011) 208–219. doi:10.1016/j.jelekin.2010.10.006.

[2]      Z.A. Riley, A.H. Maerz, J.C. Litsey, R.M. Enoka, Motor unit recruitment in human biceps brachii during sustained voluntary contractions., J. Physiol. 586 (2008) 2183–93. doi:10.1113/jphysiol.2008.150698.

[3]      D. Borzelli, M. Gazzoni, A. Botter, L. Gastaldi, A. D’Avella, T.M. Vieira, Contraction level, but not force direction or wrist position, affects the spatial distribution of motor unit recruitment in the biceps brachii muscle., Eur. J. Appl. Physiol. (2020). doi:10.1007/s00421-020-04324-6.

[4]      H.V.H.V. Cabral, L.M.L.L.M.L. de Souza, R.G.T.R.G.T. Mello, A. Gallina, L.F.L.F. de Oliveira, T.M.T.M. Vieira, Is the firing rate of motor units in different vastus medialis regions modulated similarly during isometric contractions?, Muscle and Nerve. 57 (2018) 279–286. doi:10.1002/mus.25688.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

Many thanks to the authors for revising the manuscript based on my comments. After a careful review of your revised manuscript, I think the manuscript is suitable for publication.

Author Response

Thank you very much for reviewing our revised manuscript.

Reviewer 3 Report

Authors may write a paragraph on the contribution of the article to ergonomics, sport medicine, and medicine. 

Author Response

Thank you for the suggestion. Now we have added a new paragraph at the end of the Conclusion section (Lines 426-433).

Reviewer 4 Report

Although I appreciate the authors effort in addressing my concerns, many have been left open.  In light of the authors response, I find it difficult to ascertain the validity of comparisons between force and position tasks.

My second major comments remain open.  You did not demonstrate to make like-with-like comparisons when assessing data for both force and position tasks.  I see you have used a v-shaped hinge (not shown in the Figure though) to help subjects in maintaining the requested elbow joint angle during position tasks.  Nevertheless, it was the investigator to inform the subjects whether they should pull their forearm back.  How well could the investigator discriminate fluctuations in the elbow joint angle during the task?  Forearm could have not reached the v-shaped hinge yet variations in joint angle and therefore in the firing pattern of units could have occurred, in particular for the last recruited units.  This leads to the question of whether recruitment-derecruitment data is valid for the position tasks.  The reliability results presented in Results do not contribute to addressing this issue; you just shown subjects attained similar maximal force values between visits.  Given you have not collected angle data to actually show fluctuations in joint angle were small, with small being not defined, you must provide objective data supporting the fact that subjects tracked the force template equally well in both tasks if validity of results are to be ascertained.

Still concerning this point, I am confused as to whether force was recorded during the position task.  If force was recorded in both tasks, could you not compare directly the force profiles for both force and position tasks?  You could e.g. realign the force signal of both tasks and compute the mean square error between them.  If not, why did you not record force data during position tasks?  In the absence of force data to validate your results, you must use any other supportive objective data.  The statement “the rate of increase and decrease in force during submaximal contraction was similar between the force task and position task” does not stand based on the fact that the target profile was the same between conditions.  You must show force profiles are statistically comparable between conditions

By dynamic range I was referring to the input range of you AD converter.  This information must be provided along with the amplification factor and AD resolution.

You still did not explicitly indicate how curves with the smoothed, instantaneous discharge rates (e.g. Figure 2) were computed.  How specifically did you apply the 1-s Hanning window?  Did you convolute it with a train of impulses corresponding to the discharge instants?  Why did you select 1s when others used a shorter epoch [1]?  Otherwise how did you deal with the irregular inter-pulse intervals when filtering your data? 

My concern on the definition of RT and DT was not addressed as well.  Your reply is not fully clear but it seems you have defined RT and DT as the first and last discharge instant respectively provided by the decomposition algorithm for each unit.  Users familiar with the use of decomposition algorithm to identify the firing pattern of individual units from EMGs, in particular surface EMGs, are well aware of spurious firings being regularly identified by the algorithm.  This issue is further critical if the first and last discharges are to be considered.  It is surprising then to see the authors did not apply any rigorous criteria to exclude the possibility of having defined RT and DT based on spurious discharge instants (e.g. [2,3]).  How confident readers can be the RT and DT values shown by the authors are physiologically valid?

My concern with data shown in Figure 2 was not addressed as well.  I see you have limited visualisation to the first and last two recruited units in Figure 2.  My concern however was with the roughly monotonic increase in firing rate of the last two recruited units (from ca. 2 to 8pps) without a corresponding increase in force output.  This is a rather unusual observation [4]!  Force could increase without an increase in firing rate (of some units) but the opposite is physiologically implausible.  The authors must show the spike triggered average templates for these two units, making sure these are indeed physiologically plausible units.  They must also provide compelling arguments on how the increase in firing rate for these (and possible other) units would not lead to a corresponding increase in firing rate.

[1]      C.J. de Luca, R.S. LeFever, M.P. McCue, A.P. Xenakis, Behaviour of human motor units in different muscles during linearly varying contractions, J. Physiol. 329 (1982) 113–128. doi:10.1113/jphysiol.1982.sp014293.

[2]      Z.A. Riley, A.H. Maerz, J.C. Litsey, R.M. Enoka, Motor unit recruitment in human biceps brachii during sustained voluntary contractions., J. Physiol. 586 (2008) 2183–93. doi:10.1113/jphysiol.2008.150698.

[3]      P.M. Kennedy, A.G. Cresswell, The effect of muscle length on motor-unit recruitment during isometric plantar flexion in humans, Exp Brain Res. 137 (2001) 58–64. doi:10.1007/s002210000623.

[4]      C.J. Heckman, R.M. Enoka, Motor unit, Compr. Physiol. 2 (2012) 2629–2682. doi:10.1002/cphy.c100087.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf