Ultrasound Measurement of Plantar Fascia Thickness: A Learning Curve Analysis Between a Novice and an Expert Examiner

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Thank you for the opportunity to review your manuscript, "Differences in Ultrasound Measurement of the Plantar Fascia Between Expert and Student Examiners: A Cross-Sectional Descriptive Observational Study".

The primary objective of your research—evaluating the learning curve and measurement consistency of a novice student compared to an experienced sonographer when assessing plantar fasciopathy —addresses a highly relevant clinical question. Given the widespread use of musculoskeletal ultrasound as a diagnostic tool, defining the training parameters and learning trajectories required for accurate evaluation is a practical and valuable pursuit for clinical education.

However, while the clinical premise is commendable, a careful review of the manuscript reveals several fundamental methodological and statistical issues that currently compromise the validity of your conclusions. Specifically, the statistical approach used to determine the exact "learning threshold" relies on a misinterpretation of significance testing. Furthermore, the study design lacks essential reliability metrics (such as the Intraclass Correlation Coefficient) and does not adequately control for potential confounders, such as order bias and clinical blinding.

The detailed observations provided below are offered constructively. Addressing these critical flaws is essential to strengthen the scientific rigor of your data analysis and ensure that your conclusions are fully supported by the evidence.

0.- Abstract.

The abstract exhibits a clear lack of proofreading. It includes misspellings such as "measuraments" and incorrect phrasing like "between the students and the experts' measurements" , using the plural "students" despite the methodology stating there is only one novice student. It presents a first "Conclusions" heading that unnecessarily repeats the Results section almost verbatim regarding the 0.026 p-value , which is then immediately followed by a second "Conclusion" heading

1.- Introduction

The introduction should be streamlined to focus more quickly on the specific variable being studied: ultrasound measurement techniques. The authors fail to review existing literature regarding ultrasound education, learning curves, or prior comparisons between expert and novice examiners in musculoskeletal sonography. References are inconsistents and in line 99 is not in superindex.

2.- Material and methods

The methodology suffers from several critical flaws that severely compromise its internal validity, beginning with the absence of an a priori sample size calculation, which raises concerns about potential Type II errors. Furthermore, failing to randomize the measurement sequence—since the student always performs the assessment first—introduces a systemic order bias. This issue is exacerbated by inadequate blinding regarding the patients' clinical status, creating a high risk of confirmation bias during the examinations. Additionally, using an initial ultrasound measurement (>4.0 mm) as a diagnostic inclusion criterion introduces circularity that confounds the very dependent variable being studied. Finally, the explicit omission of true agreement metrics, such as the Intraclass Correlation Coefficient (ICC), renders the statistical design fundamentally incapable of properly evaluating a learning curve or measurement consistency between observers.

3.- Results

The Results section is fundamentally invalidated by its reliance on null hypothesis significance testing to inappropriately claim measurement equivalence. By interpreting non-significant p-values (p > 0.05) as evidence that measurement differences "were no longer consistently observed," the authors commit a critical statistical fallacy. Furthermore, the sequential t-tests used to pinpoint exact "learning thresholds" at precisely 13 or 35 measurements are conducted without any correction for multiple comparisons, leading to severe Type I error inflation and mathematically unsound conclusions. The presentation of the data is also inconsistent; while some tables provide adequate detail, critical Table 6 omits means, standard deviations, and effect sizes entirely. Finally, the analysis of intraobserver variability relies solely on observational differences in standard deviations rather than employing the necessary formal reliability metrics, such as the Intraclass Correlation Coefficient, rendering the claims about measurement stability unsupported.

4.- Discussion

The Discussion section fundamentally rests on the statistical fallacies established earlier in the manuscript, erroneously conflating a non-significant p-value with the achievement of measurement accuracy and a definitive "learning threshold". While the authors appropriately acknowledge severe limitations they contradict this modesty by drawing highly specific, unwarranted conclusions, such as dictating that exactly 13 or 35 measurements are needed for competency. Furthermore, the section introduces foundational literature regarding ultrasound learning curves that should have been presented in the Introduction to properly frame the study's rationale. Finally, the authors offer highly speculative rationalizations for the variance observed in the healthy control group without empirical support, ultimately overstating the clinical and educational utility of their findings given the study's profound methodological constraints.

5.- Conclusion

The conclusions improperly reiterate the unsupported claim that supervised practice reduces measurement variability, relying entirely on the study's flawed interpretation of non-significant p-values. Although the authors acknowledge their findings lack formal reliability metrics, this brief disclaimer fails to rectify the mathematically unsound training thresholds presented as the study's core outcome.

Author Response

Reviewer 1

Thank you for the opportunity to review your manuscript, "Differences in Ultrasound Measurement of the Plantar Fascia Between Expert and Student Examiners: A Cross-Sectional Descriptive Observational Study".

The primary objective of your research—evaluating the learning curve and measurement consistency of a novice student compared to an experienced sonographer when assessing plantar fasciopathy —addresses a highly relevant clinical question. Given the widespread use of musculoskeletal ultrasound as a diagnostic tool, defining the training parameters and learning trajectories required for accurate evaluation is a practical and valuable pursuit for clinical education.

However, while the clinical premise is commendable, a careful review of the manuscript reveals several fundamental methodological and statistical issues that currently compromise the validity of your conclusions. Specifically, the statistical approach used to determine the exact "learning threshold" relies on a misinterpretation of significance testing. Furthermore, the study design lacks essential reliability metrics (such as the Intraclass Correlation Coefficient) and does not adequately control for potential confounders, such as order bias and clinical blinding.

The detailed observations provided below are offered constructively. Addressing these critical flaws is essential to strengthen the scientific rigor of your data analysis and ensure that your conclusions are fully supported by the evidence.

 

Dear Reviewer,

We sincerely thank you for your thorough and constructive evaluation of our manuscript. We appreciate your recognition of the clinical relevance of our study and the importance of understanding learning processes in musculoskeletal ultrasound.

We also acknowledge the methodological and statistical concerns raised. In response, we have carefully revised the manuscript to address these points. Specifically, we have refined the statistical analysis and its interpretation to avoid inappropriate conclusions based on significance testing, incorporated Intraclass Correlation Coefficients (ICC) to better assess measurement agreement, and clarified key methodological aspects, including measurement order, blinding considerations, and diagnostic criteria.

In addition, we have revised the Results, Discussion, and Abstract sections to ensure a more accurate and cautious interpretation of the findings, reflecting the exploratory nature of the study.

Below, we provide a detailed, point-by-point response to each of your comments.

0.- Abstract.

The abstract exhibits a clear lack of proofreading. It includes misspellings such as "measuraments" and incorrect phrasing like "between the students and the experts' measurements" , using the plural "students" despite the methodology stating there is only one novice student. It presents a first "Conclusions" heading that unnecessarily repeats the Results section almost verbatim regarding the 0.026 p-value , which is then immediately followed by a second "Conclusion" heading

We thank the reviewer for highlighting the issues related to clarity, structure, and language in the abstract.

The abstract has been carefully revised to correct typographical errors and improve grammatical accuracy. Specifically, misspellings such as “measuraments” have been corrected, and the inappropriate use of plural forms (e.g., “students”) has been revised to accurately reflect the study design involving a single novice examiner.

The structure of the abstract has also been improved to follow a standard format. The duplicated “Conclusions” section has been removed, and the content has been reorganized to eliminate redundancy and improve overall clarity.

Additionally, statements in the Conclusions section that repeated the Results—such as restating p-values—have been removed. The Conclusions have been rewritten to provide a concise interpretation of the findings rather than repeating statistical outcomes.

Finally, the abstract has been aligned with the revised manuscript to ensure consistent terminology and appropriate interpretation of results, emphasizing the exploratory nature of the findings and avoiding overstatement of measurement equivalence or reliability.

 

We believe that these revisions substantially improve the clarity, accuracy, and overall quality of the abstract.

1.- Introduction

The introduction should be streamlined to focus more quickly on the specific variable being studied: ultrasound measurement techniques. The authors fail to review existing literature regarding ultrasound education, learning curves, or prior comparisons between expert and novice examiners in musculoskeletal sonography. References are inconsistents and in line 99 is not in superindex.

We thank the reviewer for these valuable observations. The Introduction section has been revised and streamlined to focus more specifically on ultrasound measurement techniques and their operator-dependent nature, with particular emphasis on quantitative assessment of plantar fascia thickness.

In addition, relevant literature on ultrasound education, learning curves, and comparisons between novice and expert examiners in musculoskeletal sonography has been incorporated to strengthen the scientific rationale of the study. These additions provide a clearer context for understanding how measurement performance may evolve with experience and supervised training.

Furthermore, all references and citation formats have been carefully reviewed and standardized throughout the manuscript to ensure consistency. The formatting issue identified in line 99 has also been corrected, and all citations have been adjusted to the appropriate superscript format.

2.- Material and methods

The methodology suffers from several critical flaws that severely compromise its internal validity, beginning with the absence of an a priori sample size calculation, which raises concerns about potential Type II errors. Furthermore, failing to randomize the measurement sequence—since the student always performs the assessment first—introduces a systemic order bias. This issue is exacerbated by inadequate blinding regarding the patients' clinical status, creating a high risk of confirmation bias during the examinations. Additionally, using an initial ultrasound measurement (>4.0 mm) as a diagnostic inclusion criterion introduces circularity that confounds the very dependent variable being studied. Finally, the explicit omission of true agreement metrics, such as the Intraclass Correlation Coefficient (ICC), renders the statistical design fundamentally incapable of properly evaluating a learning curve or measurement consistency between observers.

We sincerely thank the reviewer for this detailed and insightful evaluation of the methodological aspects of our study. We have carefully revised the manuscript to address each of the concerns raised.

First, regarding the absence of an a priori sample size calculation, we acknowledge that this represents a limitation. The study was designed as an exploratory investigation aimed at describing learning-related changes over time rather than testing a predefined hypothesis. This has now been explicitly stated in the limitations section of the manuscript.

Second, concerning potential order bias, we clarify that the order of measurements between the student and the expert examiner was performed in a random manner to minimize potential order bias. This clarification has been incorporated into the Methods section of the manuscript.

Third, with respect to blinding, we recognize that participants’ clinical status could not be fully masked due to the nature of the condition being studied. This limitation has now been explicitly acknowledged in the revised manuscript, and interpretations have been adjusted accordingly.

Fourth, regarding the concern about circularity, we clarify that ultrasound was not used as the sole diagnostic criterion. Clinical assessment was the primary basis for classification, while ultrasound measurements were used as a complementary tool. The wording in the Methods section has been revised to reflect this more clearly.

Finally, in response to the recommendation to include formal agreement metrics, Intraclass Correlation Coefficients (ICC) have now been incorporated into the analysis. Interobserver and intraobserver ICC values are presented in the Results section and interpreted in the Discussion. These measures complement the descriptive analysis of changes over time.

We believe that these revisions substantially improve the methodological transparency and rigor of the study.

 

3.- Results

The Results section is fundamentally invalidated by its reliance on null hypothesis significance testing to inappropriately claim measurement equivalence. By interpreting non-significant p-values (p > 0.05) as evidence that measurement differences "were no longer consistently observed," the authors commit a critical statistical fallacy. Furthermore, the sequential t-tests used to pinpoint exact "learning thresholds" at precisely 13 or 35 measurements are conducted without any correction for multiple comparisons, leading to severe Type I error inflation and mathematically unsound conclusions. The presentation of the data is also inconsistent; while some tables provide adequate detail, critical Table 6 omits means, standard deviations, and effect sizes entirely. Finally, the analysis of intraobserver variability relies solely on observational differences in standard deviations rather than employing the necessary formal reliability metrics, such as the Intraclass Correlation Coefficient, rendering the claims about measurement stability unsupported.

We thank the reviewer for these valuable and insightful comments regarding the statistical interpretation of the Results section.

First, we agree that non-significant p-values should not be interpreted as evidence of equivalence. In response, we have revised the manuscript to avoid such interpretations. Expressions suggesting equivalence have been replaced by more appropriate wording, such as “differences were no longer consistently detected” or “differences decreased over time,” ensuring a descriptive rather than confirmatory interpretation.

Second, we acknowledge the reviewer’s concern regarding the identification of specific learning thresholds (e.g., 13 or 35 measurements). We have revised the manuscript to remove deterministic statements and now present these findings as indicative patterns rather than exact thresholds. The revised text emphasizes that these results suggest a progressive reduction in measurement differences associated with increased experience, rather than defining precise competency cut-offs.

Third, regarding multiple comparisons, we clarify that the sequential analyses were exploratory in nature. This has now been explicitly stated in both the Results and Discussion sections, and interpretations have been appropriately moderated to reflect this limitation.

Fourth, Table 6 has been improved to enhance clarity and interpretability. Additional explanatory text has been included to better contextualize the sequential p-values and their role in illustrating trends over time.

Finally, in response to the comment regarding intraobserver variability, Intraclass Correlation Coefficients (ICC) have now been incorporated into the analysis. These provide a formal measure of agreement and complement the descriptive findings. Interobserver and intraobserver ICC values have been reported and interpreted accordingly.

We believe these revisions significantly improve the methodological rigor and clarity of the Results section.

 

4.- Discussion

The Discussion section fundamentally rests on the statistical fallacies established earlier in the manuscript, erroneously conflating a non-significant p-value with the achievement of measurement accuracy and a definitive "learning threshold". While the authors appropriately acknowledge severe limitations they contradict this modesty by drawing highly specific, unwarranted conclusions, such as dictating that exactly 13 or 35 measurements are needed for competency. Furthermore, the section introduces foundational literature regarding ultrasound learning curves that should have been presented in the Introduction to properly frame the study's rationale. Finally, the authors offer highly speculative rationalizations for the variance observed in the healthy control group without empirical support, ultimately overstating the clinical and educational utility of their findings given the study's profound methodological constraints.

We thank the reviewer for these important observations regarding the Discussion section.

We fully agree that non-significant p-values should not be interpreted as evidence of measurement accuracy or equivalence. Accordingly, we have revised the Discussion to ensure that findings are described in terms of trends and patterns rather than definitive conclusions. Statements suggesting equivalence have been replaced with more cautious language reflecting reduced or attenuated differences.

Regarding the identification of specific learning thresholds (13 and 35 measurements), we have modified the discussion to avoid presenting these values as precise or prescriptive. They are now described as exploratory observations that may reflect a gradual reduction in measurement differences rather than clearly defined competency thresholds.

We have also clarified that all sequential analyses should be interpreted with caution due to their exploratory nature and the absence of correction for multiple comparisons.

The discussion of learning curves and relevant literature has been refined and more clearly linked to the study’s aims. These references are now used to contextualize the findings rather than to support definitive claims.

Finally, speculative interpretations—particularly regarding variability in healthy fascia—have been revised to reflect a more cautious and evidence-informed approach.

Overall, the Discussion has been substantially revised to align interpretation with the exploratory design of the study and to avoid overstatement of the findings.

 

5.- Conclusion

The conclusions improperly reiterate the unsupported claim that supervised practice reduces measurement variability, relying entirely on the study's flawed interpretation of non-significant p-values. Although the authors acknowledge their findings lack formal reliability metrics, this brief disclaimer fails to rectify the mathematically unsound training thresholds presented as the study's core outcome.

We appreciate the reviewer’s comments regarding the Conclusion section.

In response, the conclusions have been revised to ensure that they are consistent with the exploratory nature of the study. Strong causal statements regarding the reduction of variability have been moderated, and language suggesting definitive effects has been replaced with more cautious expressions such as “may help improve” or “suggests.”

Additionally, references to specific measurement thresholds have been removed from the conclusions to avoid overinterpretation of the sequential analyses.

The revised conclusion emphasizes that the findings provide descriptive evidence of a learning effect, rather than formal proof of interobserver reliability. The importance of structured training is now presented as a potential implication rather than a definitive outcome.

Finally, we believe that these revisions have substantially improved the overall quality and clarity of the manuscript, and we hope that the revised version will be considered suitable for publication.

Thank you again for your time and valuable feedback.

Reviewer 2 Report

Comments and Suggestions for Authors

This manuscript addresses a clinically relevant and practically important question regarding US training and operator dependent variability in plantar fascia assessment. Topic is interesting and potentially valuable for musculoskeletal US evaluators.

1. The main concern is that the authors investigate measurement agreement without applying formal agreement methodology. The analyses rely primarily on p values and effect sizes to compare student and expert measurements but statistical non-significance cannot be interpreted as agreement! If the aim is to evaluate interobserver or intraobserver performance, more appropriate methods such as ICC. Although the authors repeatedly state that this is not a reliability study, several interpretations still imply convergence of measurements and improved measurement consistency beyond what the analyses support.
2. A second concern relates to the proposed learning threshold. The conclusion that approximately 13 pathological measurements or 35 total measurements are sufficient before differences disappear is overstating. These thresholds seem to result from sequential repeated testing without predefined criteria or adjustment for multiple comparisons. Also, conclusions are based on a single student and single expert.. These findings should therefore be presented as exploratory rather than estimates of training requirements or learning curves!
3. One smaller issue is regardfing definition of plantar fasciopathy. Manuscript combines symptoms, clinical examination and ultrasound thickness criteria during participant classification. Since ultrasound thickness is also the primary outcome, the authors should clarify whether ultrasound findings were part of inclusion criteria or only used for descriptive characterization to avoid circular reasoning.
4. Discussion is weak in its relevance to musculoskeletal radiology practice. The authors should compare their findings with similar studies evaluating ultrasUSound learning curves, interobserver performance, and training requirements, with particular emphasis on whether structured education improves measurement accuracy and how much training may be required to achieve consistent interpretation. Expanding the reference list to include additional studies on ultrasound education and musculoskeletal imaging training would strengthen the clinical and educational implications of the manuscript.

Author Response

Response reviewer 2

This manuscript addresses a clinically relevant and practically important question regarding US training and operator dependent variability in plantar fascia assessment. Topic is interesting and potentially valuable for musculoskeletal US evaluators.

1. The main concern is that the authors investigate measurement agreement without applying formal agreement methodology. The analyses rely primarily on p values and effect sizes to compare student and expert measurements but statistical non-significance cannot be interpreted as agreement! If the aim is to evaluate interobserver or intraobserver performance, more appropriate methods such as ICC. Although the authors repeatedly state that this is not a reliability study, several interpretations still imply convergence of measurements and improved measurement consistency beyond what the analyses support.

Dear Reviewer,

We sincerely thank you for your thorough and constructive evaluation and for this important comment regarding the assessment of agreement.

We fully agree that non-significant p-values should not be interpreted as evidence of agreement. In response, the manuscript has been revised to eliminate such interpretations. Expressions implying equivalence have been replaced with more appropriate language, including “differences were no longer consistently detected” and “differences decreased over time,” ensuring a descriptive rather than confirmatory interpretation of the findings.

In addition, Intraclass Correlation Coefficients (ICC) have now been incorporated into the analysis to provide a formal measure of agreement. Both interobserver and intraobserver ICC values are reported in the Results section and further interpreted in the Discussion.

These measures complement the descriptive analysis and allow a more appropriate evaluation of measurement consistency without changing the exploratory nature of the study.

 

2. A second concern relates to the proposed learning threshold. The conclusion that approximately 13 pathological measurements or 35 total measurements are sufficient before differences disappear is overstating. These thresholds seem to result from sequential repeated testing without predefined criteria or adjustment for multiple comparisons. Also, conclusions are based on a single student and single expert.These findings should therefore be presented as exploratory rather than estimates of training requirements or learning curves!

We appreciate the reviewer’s concern regarding the interpretation of learning thresholds.

We agree that the values reported (13 and 35 measurements) should not be interpreted as precise or definitive thresholds. The manuscript has been revised to avoid deterministic statements, and these findings are now presented as exploratory patterns suggesting a progressive reduction in measurement differences with increasing experience.

We have also clarified that the sequential analyses were exploratory in nature and involved multiple comparisons, and therefore should be interpreted with caution.

Furthermore, we explicitly acknowledge that the study included only one novice examiner and one expert examiner, which limits generalizability. The findings are therefore presented as descriptive observations rather than estimates of training requirements or learning curves applicable to broader populations.

 

3. One smaller issue is regardfing definition of plantar fasciopathy. Manuscript combines symptoms, clinical examination and ultrasound thickness criteria during participant classification. Since ultrasound thickness is also the primary outcome, the authors should clarify whether ultrasound findings were part of inclusion criteria or only used for descriptive characterization to avoid circular reasoning.

We thank the reviewer for this important methodological observation.

We clarify that ultrasound was not used as the sole diagnostic criterion for participant classification. Clinical assessment was the primary basis for group allocation, while ultrasound measurements were used as a complementary tool for characterization.

The wording in the Methods section has been revised to clearly reflect this point and avoid potential concerns regarding circular reasoning.

 

4. Discussion is weak in its relevance to musculoskeletal radiology practice. The authors should compare their findings with similar studies evaluating ultrasUSound learning curves, interobserver performance, and training requirements, with particular emphasis on whether structured education improves measurement accuracy and how much training may be required to achieve consistent interpretation. Expanding the reference list to include additional studies on ultrasound education and musculoskeletal imaging training would strengthen the clinical and educational implications of the manuscript

We thank the reviewer for this valuable suggestion.

The Discussion has been revised to strengthen its relevance to musculoskeletal ultrasound practice. We have expanded the integration of existing literature on ultrasound learning curves, interobserver performance, and training in musculoskeletal imaging.

Specifically, we now compare our findings with previous studies evaluating learning-related changes in ultrasound performance and discuss how structured training and supervised practice may influence measurement consistency.

Furthermore, the Introduction has been revised to include additional references on ultrasound education and expert–novice comparisons, providing a more robust rationale for the study.

 

These additions improve the clinical and educational relevance of the manuscript.

Finally, we believe that these revisions have substantially improved the overall quality and clarity of the manuscript, and we hope that the revised version will be considered suitable for publication.

Thank you again for your time and valuable feedback.

 

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

Dear Authors, I would like to sincerely thank you for your efforts in preparing this revised version of the manuscript and for carefully addressing all the comments and suggestions raised during the previous review round. After a thorough evaluation of the updated document, I can confirm that there are no remaining methodological, statistical, or content errors, and that the chosen two-way mixed-effects absolute agreement model (ICC(3,1)) remains highly appropriate for your study design. The modifications introduced have significantly enhanced the clarity of the text, further strengthening an already excellent approach to the learning curve in the ultrasound measurement of the plantar fascia, particularly regarding your brilliant discussion on why healthy fascia poses a greater perceptual challenge for beginners. Therefore, I consider that the manuscript has successfully reached the required level of scientific quality and is now fully suitable for publication, subject only to a few final, minor editorial corrections before typesetting. Congratulations on an excellent and very practical piece of work.

Author Response

Dear Reviewer,

We sincerely thank you for your careful and thorough evaluation of our revised manuscript. We deeply appreciate your kind and encouraging comments, as well as your recognition of the improvements made in response to the previous review.

We would also like to inform you that an updated version of the manuscript, including the minor editorial corrections suggested by Reviewer 2, has now been submitted.

Your insightful feedback has been invaluable throughout this process, and we truly appreciate your support in bringing this work to its final stage.

Thank you again for your time and constructive input.

Kind regards,

Mª Carmen Ledesma

Corresponding author

 

Reviewer 2 Report

Comments and Suggestions for Authors

I appreciate that the authors have added ICC analyses and have substantially improved the interpretation of their findings. However, I would still encourage a more cautious interpretation of the temporal ICC results. Although ICC values were numerically higher in later periods (0,67, 0,73, 0,80) no formal comparison between ICC estimates was performed and the confidence intervals overlap considerably. Therefore statements suggesting that interobserver agreement progressively increased over time should be presented as descriptive observatinos rather than demonstrated improvements in agreement..

The expanded discussion on ultrasound learning curves and the importance of supervised training was added but somehow needs to be further strengthened. Authors do not comment that operator experience influences not only measurement reproducibility but also the correct interpretation of ultrasound images. Previous studies have demonstrated that less experienced examiners are more prone to misidentifying anatomical structures and imaging artifacts, whereas greater proficiency improves discrimination between true tissue features and artefactual findings. This aspect may be particularly relevant in musculoskeletal ultrasound, where recognition of anisotropy and other artefacts is essential for accurate assessment of structures such as tendons and fascia (see, comment: https://www.jhandsurg.org/article/S0363-5023(26)00309-6/fulltext)

The clarification regarding measurement order ca be also added. The statement that “the order was randomized but occasionally followed logistical constraints” remains somewhat unclear?? Additional information on the extent to which the intended randomization was implemented would help readers assess the potential for bias..

Author Response

 

Dear Reviewer,

We would like to sincerely thank you for your careful review and for the valuable comments and suggestions provided on our manuscript. We truly appreciate the time and effort you have devoted to helping us improve the quality of our work.

We have addressed each of your comments in detail, and our point-by-point responses can be found in the attached document.

Thank you again for your constructive feedback.

Kind regards,

Mª Carmen Ledesma 

Corresponding author

 

 

Author Response File: Author Response.pdf

Round 3

Reviewer 2 Report

Comments and Suggestions for Authors

Authors improved overall manuscript quality, particularly the discussion has been enhanced with more balancing and cautious interpretation of ICC data. I think the manuscript is now in appropriate form to be endorsed for publication.