The Influence of Menstrual Cycle Phase and Urinary Incontinence on Potential ACL Injury Risk Factors with a Focus on Hip Strength and Postural Control in Elite Female Team Sport Athletes: A Pilot Study

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors Overall this is an interesting study and I believe the readers of the Journal will find it informative. For the most part it is clear and well written. Minor suggestions below. -This isn't really a study of ACL risk factors aside from abd:add hip strength.  Even as the authors mentioned, there was not an analysis of movement tasks that are associated with ACL injuries (i.e., cutting, decel).  The authors should consider rethinking the title of the study to be the relationship between ab/ad strength and UI/MC phases.  -The second cannonical relationship described between well-being and add, sway path, and sway area isn't well described in terms of translation or relevance.  In the discussion the MC symptoms and performance being variables are discussed, but this particular result isn't addressed.  If there is a significant relationship, it doesn't seem clear what the importance of that is.  This warrants further discussion.  -Figure 4 is crowded and difficult to discern.  Consider choosing the most impactful to embed in the paper and move the other figures into a supplemental document to make it easier to read. 

Author Response

Comments and Suggestions for Authors

Overall this is an interesting study and I believe the readers of the Journal will find it informative. For the most part it is clear and well written. Minor suggestions below.

Authors response: We sincerely thank the reviewer for their constructive and thoughtful feedback. Their comments have helped us clarify the study’s focus, improve the presentation of our results, and enhance the clarity and readability of the manuscript.

 

• This isn't really a study of ACL risk factors aside from abd:add hip strength.  Even as the authors mentioned, there was not an analysis of movement tasks that are associated with ACL injuries (i.e., cutting, decel).  The authors should consider rethinking the title of the study to be the relationship between ab/ad strength and UI/MC phases. 

Authors response: We agree and have revised the title for clarity. It now reads: “Influence of menstrual cycle phase and urinary incontinence on potential ACL injury risk factors with a focus on hip strength and postural control in elite female team sport athletes: a pilot study.”

 

• The second cannonical relationship described between well-being and add, sway path, and sway area isn't well described in terms of translation or relevance.  In the discussion the MC symptoms and performance being variables are discussed, but this particular result isn't addressed.  If there is a significant relationship, it doesn't seem clear what the importance of that is.  This warrants further discussion. 

Authors response: According to your suggestion, we have addressed this point in the Limitations section by noting that, although some canonical correlations were statistically significant, certain coefficients and structure coefficients were counterintuitive: “Regarding the CCA, although statistically significant, some coefficients and structure coefficients were counterintuitive. For example, higher well-being scores (indicating worse well-being) were occasionally associated with better performance on risk factors. This is likely attributable to methodological constraints, including low variance in MC-related symptoms, which reduces the stability of canonical coefficients, and the relatively small sample size in relation to the number of variables, which can affect the reliability of the canonical functions.”

 

• Figure 4 is crowded and difficult to discern.  Consider choosing the most impactful to embed in the paper and move the other figures into a supplemental document to make it easier to read. 

Authors response: We agree with the reviewer’s comment. To improve clarity and readability of the manuscript, Figure 4 has been moved to the supplemental material (Appendix A.2).

Reviewer 2 Report

Comments and Suggestions for Authors

The manuscript addresses a highly relevant and timely topic in female elite sports, integrating menstrual cycle phase determination, urinary incontinence, neuromuscular risk factors, and advanced multivariate statistics. The use of gold-standard menstrual cycle verification and the inclusion of pelvic floor considerations are notable strengths. However, the very small sample size, the complexity of the statistical models relative to n, and some overextensions in interpretation limit the robustness of the conclusions. Several sections would benefit from tighter methodological justification, clearer distinction between association and inference, and improved clarity of english expression.

Title: The title suggests a direct influence on “ACL injury risk,” while the study only evaluates ACL injury risk factors, not injury incidence. Authors should rephrase the title to explicitly reflect risk factors to avoid causal or clinical overstatement.

Abstract: The concluding sentence implies practical applicability (“optimize training and support athlete well-being”) without clearly acknowledging the limited external validity. Please temper the conclusion by explicitly referencing the pilot nature and limited generalizability.

Introduction

The idea that the follicular phase and ovulation are periods of increased ACL susceptibility is widely accepted, despite ongoing debate and inconsistencies in the literature. When introducing this claim, authors should more clearly acknowledge the inconsistency and methodological limitations of existing evidence.

 

Hip abduction strength and postural control are both described as “key risk factors”, but the strength of the evidence for each differs substantially. Authors should clearly differentiate between well-established risk factors (e.g. hip strength) and those that are more debatable or indirect (e.g. static postural control).

 

The mechanistic link between pelvic floor muscles, UI, and ACL injury risk is plausible but remains speculative. Authors should explicitly frame this section as a hypothesized pathway rather than an established mechanism.

Methods

Study design: The design is described as “prospective longitudinal,” but only one menstrual cycle contributes data to the primary analyses. Please clarify this part.

Participants:  Ten out of 24 athletes were excluded due to menstrual dysfunction, which may indicate strong selection bias. Authors should discuss how this high exclusion rate affects representativeness and external validity.

Stress urinary incontinence classification of UI using a cutoff ≥1 point on the ICIQ-UI SF includes very mild cases. Please justify this threshold more explicitly and discuss its implications for clinical relevance.

Menstrual cycle monitoring: The three-step method is well described, but the relative contribution of each step to final classification is not fully clear. Authors should clarify whether hormonal confirmation ever contradicted diary or ovulation test data.

The progesterone threshold for luteal phase deficiency has been adopted from previous studies, but the saliva-based cut-offs are still being debated. Please acknowledge the ongoing debate regarding salivary hormone thresholds briefly in the discussion.

Test protocol: The decision not to standardize warm-up may introduce uncontrolled variability across sessions and phases. Authors should more clearly justify this choice and discuss its potential impact on reliability.

Static postural control: Trial order was fixed rather than randomized, which may introduce order or fatigue effects. Please justify the fixed order or acknowledge this as a methodological limitation.

Hip strength: Only isometric strength was assessed, whereas ACL injuries typically occur during dynamic, high-velocity tasks. Please acknowledge this limitation more explicitly in the methods or discussion.

Statistical analyses

Mixed-model ANOVA with n = 10 and a between-subjects factor (UI vs. non-UI) raises concerns about statistical power and robustness. Authors should explicitly justify this analytical choice and emphasize effect sizes and uncertainty rather than p-values.

The use of multiple CCAs with many variables relative to sample size increases the risk of overfitting. Please provide a stronger rationale for CCA in this pilot context and clearly frame all multivariate findings as exploratory.

Results

Participant characteristics and table 1:  Descriptive differences between UI and non-UI groups are presented without caution regarding inferential interpretation. Authors should explicitly state that these comparisons are descriptive only.

Effects of MC phase and UI: reporting a significant main effect of UI on the adduction:abduction ratio risks causal interpretation. Authors should clearly state that this is an association, not evidence of causation.

Canonical Correlation Analysis: The percentages of explained variance may appear substantial, but they are derived from redundancy indices in a very small sample. Reiterate that these values should not be interpreted as stable or generalisable effect estimates.

 

Discussion: The discussion occasionally implies practical relevance beyond what the data can support. Please further align interpretations with the pilot design and exploratory statistics.

The mechanistic narrative linking pelvic floor coactivation, UI, balance, and strength to ACL risk is plausible, but this study did not measure pelvic floor function or neuromuscular coactivation. Please reframe these sentences as a hypothesized pathway and explicitly acknowledge that pelvic floor contribution is not directly tested here.

Recommending perturbation tasks is good, but it reads as prescriptive without acknowledging feasibility, reliability, and standardization issues across labs. Authors should add 1 or 2 sentences acknowledging practical constraints and propose concrete alternatives.

The comparison to “physically active women training twice a week” is not directly relevant to elite athletes and may distract from the main argument. Consider shortening and focusing on why static EC single-leg stance may have ceiling effects in elite athletes.

Strengths: Strengths are well stated, but they could be more balanced by acknowledging that gold-standard phase verification does not compensate for limited statistical power and model complexity. Add one sentence stating that although phase verification is robust, inferential conclusions remain limited by sample size and analytic flexibility.

They acknowledge that two cycles would be desirable; however, given the study rationale, this limitation is substantial and should be linked to within-person variability and reliability. Authors should explicitly describe how single-cycle assessment may inflate random noise and reduce the reproducibility of “phase effects”.

Limitations are well acknowledged, but the implications of CCA instability are not fully developed. Authors should more clearly state that multivariate findings are hypothesis generating only.

Conclusions: Conclusions are generally balanced but still suggest applicability to training management. Please explicitly state that recommendations are conceptual rather than evidence-based prescriptions.

 

 

Author Response

Comments and Suggestions for Authors

The manuscript addresses a highly relevant and timely topic in female elite sports, integrating menstrual cycle phase determination, urinary incontinence, neuromuscular risk factors, and advanced multivariate statistics. The use of gold-standard menstrual cycle verification and the inclusion of pelvic floor considerations are notable strengths. However, the very small sample size, the complexity of the statistical models relative to n, and some overextensions in interpretation limit the robustness of the conclusions. Several sections would benefit from tighter methodological justification, clearer distinction between association and inference, and improved clarity of english expression.

Authors response: We sincerely thank the reviewer for their careful and insightful feedback. Your thoughtful suggestions have been invaluable in helping us clarify key points, strengthen the interpretation of our findings, and more clearly communicate the exploratory nature of this pilot study. We have carefully revised the manuscript in response to your comments, including a thorough check and improvement of the English throughout. We believe these revisions have significantly enhanced the manuscript's clarity and quality.

Title: The title suggests a direct influence on “ACL injury risk,” while the study only evaluates ACL injury risk factors, not injury incidence. Authors should rephrase the title to explicitly reflect risk factors to avoid causal or clinical overstatement.

Authors response: According to your suggestion and in line with Reviewer 1’s suggestion, we have rephrased the title to avoid causal or clinical overstatement and to clearly reflect what the study examines. The revised title now reads: “Influence of menstrual cycle phase and urinary incontinence on potential ACL injury risk factors with a focus on hip strength and postural control in elite female team sport athletes: a pilot study.”

 

Abstract: The concluding sentence implies practical applicability (“optimize training and support athlete well-being”) without clearly acknowledging the limited external validity. Please temper the conclusion by explicitly referencing the pilot nature and limited generalizability.

Authors response: We agree. The concluding sentence of the abstract has been revised to acknowledge the study's pilot nature and limited generalizability. It now reads: “Although limited by its pilot design, menstrual symptoms, more than MC phases, may influence performance and injury risk, supporting the potential value of systematic symptom monitoring.”

Additionally, we have included “a pilot study” in the title to clearly indicate the exploratory nature of the research.

 

Introduction

The idea that the follicular phase and ovulation are periods of increased ACL susceptibility is widely accepted, despite ongoing debate and inconsistencies in the literature. When introducing this claim, authors should more clearly acknowledge the inconsistency and methodological limitations of existing evidence.

Authors response: To address this point, and following your suggestion, we have added a sentence in the introduction acknowledging the limitations and inconsistencies in the existing literature regarding ACL injury risk across the menstrual cycle. The added text reads: “However, evidence for increased ACL injury risk during the follicular phase and around ovulation, as well as for associated biomechanical and neuromuscular risk factors, re-mains limited due to poor-quality MC assessment and study design [5,8]. ”

 

Hip abduction strength and postural control are both described as “key risk factors”, but the strength of the evidence for each differs substantially. Authors should clearly differentiate between well-established risk factors (e.g. hip strength) and those that are more debatable or indirect (e.g. static postural control).

Authors response: We thank the reviewer for this suggestion and fully agree. The sentence has been reframed to distinguish between well-established and more tentative risk factors. It now reads: “Among (neuro-)muscular risk factors for ACL injuries, decreased hip abduction strength is well-established, whereas evidence for reduced postural control is less consistent [4,9].”

 

The mechanistic link between pelvic floor muscles, UI, and ACL injury risk is plausible but remains speculative. Authors should explicitly frame this section as a hypothesized pathway rather than an established mechanism.

Authors response: According to your suggestion we have clarified the section to emphasize that the proposed link is hypothetical rather than established. Specifically, we added the following sentence: “Taken together, these observations suggest a hypothetical pathway in which pelvic floor function, MC phase, and UI could influence ACL injury susceptibility.”

 

Methods

Study design: The design is described as “prospective longitudinal,” but only one menstrual cycle contributes data to the primary analyses. Please clarify this part.

Authors response: We agree. The study design has been clarified as follows: “A prospective longitudinal observational design was chosen to investigate the relationship between hip strength and static as well as dynamic balance across a single MC in elite female athletes.”

 

Participants:  Ten out of 24 athletes were excluded due to menstrual dysfunction, which may indicate strong selection bias. Authors should discuss how this high exclusion rate affects representativeness and external validity.

Authors response: According to your suggestion we have clarified this limitation in the manuscript: “A limitation of this pilot study is the small sample size (n = 10), which may introduce selection bias, limit the statistical power of the results and reduce their generalizability to the broader target population.” Additionally, we highlight in the discussion (4.1.) that future studies should include a broader range of athletes, encompassing those with and without menstrual dysfunction or using hormonal contraception, to better capture variability in menstrual and hormonal profiles and their potential interactions with ACL injury risk.

 

Stress urinary incontinence classification of UI using a cutoff ≥1 point on the ICIQ-UI SF includes very mild cases. Please justify this threshold more explicitly and discuss its implications for clinical relevance.

Authors response: According to your suggestion, we have clarified the classification threshold in the Methods section: “Participants were classified as having UI if they scored ≥1 point on the questionnaire [22], in order to capture any presence of symptoms, including very mild cases, that still may have an impact in the context of elite sports.” Additionally, we addressed the clinical implications in the Discussion: “Specifically, three athletes presented with mild UI (ICIQ-UI SF scores 3–4) and one athlete with moderate UI (score 6) [22], and therefore, the findings may not be generalizable to athletes with more severe UI.”

 

Menstrual cycle monitoring: The three-step method is well described, but the relative contribution of each step to final classification is not fully clear. Authors should clarify whether hormonal confirmation ever contradicted diary or ovulation test data.

Authors response: We thank the reviewer for this comment. The three-step method was conducted sequentially, not in parallel, so cases of oligomenorrhea, anovulation, and short luteal phase were excluded at the earlier steps and did not proceed to hormonal verification. Hormonal verification was therefore only performed when ovulation was confirmed, in accordance with the guidelines of Elliott-Sale et al. (https://doi.org/10.1007/s40279-021-01435-8). Saliva hormone measurements were primarily used to verify menstrual cycle phase classification and generally confirmed diary and ovulation test results. In one case, luteal phase deficiency was detectable only via hormonal verification, which led to exclusion from the analysis.

 

According to your suggestion, and to clarify the contribution of each step to the final classification and the sequential nature of the procedure, we have added the following explanation to the caption of Figure 2: “Figure 2. Flow chart of participant inclusion. Exclusion due to MC dysfunction corresponded to the step of the “three-step method” in which it was identified. In step 1, oligomenorrhea was detected via diary monitoring (n = 2), in step 2, anovulation and short luteal phase were detected via ovulation testing (n = 7), and in step 3, luteal phase deficiency was identified via hormonal verification (n = 1).”

 

The progesterone threshold for luteal phase deficiency has been adopted from previous studies, but the saliva-based cut-offs are still being debated. Please acknowledge the ongoing debate regarding salivary hormone thresholds briefly in the discussion.

Authors response: We agree and have added the following in the Discussion (4.1): “One athlete was excluded due to luteal phase deficiency. The progesterone threshold used was adopted from a previous study [26]. However, salivary cut-offs for menstrual dysfunction remain debated and require further validation.”

 

Test protocol: The decision not to standardize warm-up may introduce uncontrolled variability across sessions and phases. Authors should more clearly justify this choice and discuss its potential impact on reliability.

Authors response: According to your suggestion, we justified our decision not to standardize the warm-up. We added the following sentence to the Methods section: “Allowing athletes to perform warm-ups they perceive as most beneficial accounts for inter- and intra-individual differences, supports optimal preparation, and enables full focus on subsequent performance testing [27].”

This approach acknowledges that athletes’ responses to a standardized warm-up can vary, and giving them the flexibility to modulate their routine based on previous experience, daily readiness, and individual preferences helps ensure consistent performance despite differences in warm-up content.

 

Static postural control: Trial order was fixed rather than randomized, which may introduce order or fatigue effects. Please justify the fixed order or acknowledge this as a methodological limitation.

Authors response: The procedure was oriented according to the protocol by Fridén et al. (https://doi.org/10.1159/000086592), which allows standardization and facilitates comparison with previous studies. According to your suggestion, we have added the following sentence to the Methods section to justify the fixed order: “This fixed order standardized progression from easier to more challenging stances and ensured participant safety.”

 

Hip strength: Only isometric strength was assessed, whereas ACL injuries typically occur during dynamic, high-velocity tasks. Please acknowledge this limitation more explicitly in the methods or discussion.

Authors response: We thank the reviewer for pointing this out and fully agree. We have revised the discussion (section 4.2) to clarify this by adding “Similarly, hip strength was assessed only isometrically, which may not fully capture dynamic, high-velocity movements that typically precede ACL injuries.”

 

Statistical analyses

Mixed-model ANOVA with n = 10 and a between-subjects factor (UI vs. non-UI) raises concerns about statistical power and robustness. Authors should explicitly justify this analytical choice and emphasize effect sizes and uncertainty rather than p-values.

Authors response: We agree with the reviewer that the use of mixed-model ANOVA with a small sample size and a between-subjects factor limits statistical power and robustness. To address this, we have explicitly acknowledged the exploratory nature of these analyses in the Methods section by adding the following statement: “Due to the small sample size, the following ANOVA results should be interpreted as exploratory. Statistical significance is reported to aid interpretation of observed association patterns but should not be considered confirmatory.”

In addition, effect sizes have been added throughout the Results section to provide a clearer representation of the magnitude of observed effects.

 

The use of multiple CCAs with many variables relative to sample size increases the risk of overfitting. Please provide a stronger rationale for CCA in this pilot context and clearly frame all multivariate findings as exploratory.

Authors response: We agree with the reviewer that the use of multiple CCAs with a small sample size increases the risk of overfitting. To address this, we have strengthened the rationale for using CCA and clearly framed all multivariate findings as exploratory. Specifically, we added the following statement to the Methods section: “A CCA was chosen to explore multivariate association patterns between perceptual/symptom-related variables and neuromuscular and postural control measures, as these domains are theoretically interdependent and unlikely to be adequately represented by isolated univariate or bivariate analyses, which may have further inflated the number of comparisons and the likelihood of type-I errors. Despite the small sample size, CCA was deemed suitable for hypothesis-generating purposes, allowing the identification of shared variance structures across variable sets that may inform future confirmatory studies.”

 

Results

Participant characteristics and table 1:  Descriptive differences between UI and non-UI groups are presented without caution regarding inferential interpretation. Authors should explicitly state that these comparisons are descriptive only.

Authors response: We agreed with the reviewer. To avoid any inferential interpretation, we revised the caption of Table 1 to read “Descriptive characteristics of the final participant group (n = 10)”

 

Effects of MC phase and UI: reporting a significant main effect of UI on the adduction:abduction ratio risks causal interpretation. Authors should clearly state that this is an association, not evidence of causation.

Authors response: According to your suggestion the text has been revised to emphasize association rather than causation. It now reads: “Additionally, the between-subjects factor UI was significantly associated with the adduction:abduction-ratio of hip strength, with higher values observed in the UI group (p = 0.037).“

 

Canonical Correlation Analysis: The percentages of explained variance may appear substantial, but they are derived from redundancy indices in a very small sample. Reiterate that these values should not be interpreted as stable or generalisable effect estimates.

Authors response: We agree. To address this, we added the following clarification to the Methods section: “Given the small final sample size, all statistical analyses were conducted with an awareness of limited statistical power and an increased risk of Type I error. Consequently, the mixed-model ANOVA and CCA were intended to explore multivariate association patterns rather than to provide confirmatory inference.“

Additionally, we emphasized this point in the Limitations section: “Given the small sample size, the present findings, especially results of the multivariate analyses, should be considered hypothesis-generating. The identified multivariate association patterns warrant replication in larger, adequately powered cohorts before firm conclusions can be drawn.“

 

Discussion: The discussion occasionally implies practical relevance beyond what the data can support. Please further align interpretations with the pilot design and exploratory statistics.

Authors response: We thank the reviewer for this comment. According to your suggestion, the Discussion has been carefully revised to align interpretations with the pilot design and exploratory nature of the analyses. We have added multiple references to e.g., “pilot study” throughout and included the following statement at the beginning of the Discussion: “As a pilot study with a small sample of elite female athletes, the present findings are exploratory and intended to generate hypotheses rather than provide definitive conclusions about ACL injury risk factors across MC phases.”

 

The mechanistic narrative linking pelvic floor coactivation, UI, balance, and strength to ACL risk is plausible, but this study did not measure pelvic floor function or neuromuscular coactivation. Please reframe these sentences as a hypothesized pathway and explicitly acknowledge that pelvic floor contribution is not directly tested here.

Authors response: In line with your and Reviewer 3’s suggestion, we have reframed the discussion of pelvic floor muscles as a hypothesized pathway. Specifically, we added: “Further UI in this study was assessed indirectly using the ICIQ-UI SF questionnaire. Although widely used and validated [21], this instrument does not provide a direct, objective evaluation of pelvic floor muscle function. More detailed and objective evaluations of pelvic floor muscle function, such as palpation or sonographic assessment, could yield a more accurate characterization of pelvic floor muscles [38,39]. It is plausible that more pronounced differences would emerge in athletes with more severe UI or when pelvic floor function is assessed directly, including across multiple MC phases rather than a single time point.”

 

Recommending perturbation tasks is good, but it reads as prescriptive without acknowledging feasibility, reliability, and standardization issues across labs. Authors should add 1 or 2 sentences acknowledging practical constraints and propose concrete alternatives.

Authors response: We agree and have added the following to the Discussion: “Perturbation-based balance tasks could provide a more valid assessment of neuromuscular control by requiring active recovery of stability. However, such approaches have practical limitations: oscillatory platforms are typically laboratory-based, and perturbations are generally difficult to standardize across participants and sessions. As a feasible alternative, future research could focus on whole-body kinematics during dynamic movements, such as side-cutting or change-of-direction tasks, which may provide informative, field-applicable insights into neuromuscular control in elite athletes [41]. Overall, this suggests that both static balance tests and isometric strength assessments may not fully capture the dynamic neuromuscular demands relevant to ACL injury risk in elite female athletes, highlighting the need for protocols that reflect the neuromuscular demands of elite athletes.”

 

The comparison to “physically active women training twice a week” is not directly relevant to elite athletes and may distract from the main argument. Consider shortening and focusing on why static EC single-leg stance may have ceiling effects in elite athletes.

Authors response: We agree. To maintain focus on elite athletes, we have shortened the text by removing the comparison to recreationally active women.

 

Strengths: Strengths are well stated, but they could be more balanced by acknowledging that gold-standard phase verification does not compensate for limited statistical power and model complexity. Add one sentence stating that although phase verification is robust, inferential conclusions remain limited by sample size and analytic flexibility.

Authors response: According to your suggestion, we have added the following sentence to the end of the strengths section: “Although MC phase verification was robust, inferential conclusions remain limited by the small sample size and the complexity of the multivariate analyses.”

 

They acknowledge that two cycles would be desirable; however, given the study rationale, this limitation is substantial and should be linked to within-person variability and reliability. Authors should explicitly describe how single-cycle assessment may inflate random noise and reduce the reproducibility of “phase effects”.

Authors response: Thank you for pointing this out. In accordance with your suggestion, we have added the following clarification to the Limitations section: “Evaluating only one MC may introduce additional random variation and limit the reproducibility of phase-related findings due to within-person variability.”

 

Limitations are well acknowledged, but the implications of CCA instability are not fully developed. Authors should more clearly state that multivariate findings are hypothesis generating only.

Authors response: We agree and have added the following clarification to the Limitations section: “Given the small sample size, the present findings, especially results of the multivariate analyses, should be considered hypothesis-generating. The identified multivariate association patterns warrant replication in larger, adequately powered cohorts before firm conclusions can be drawn.“

 

Conclusions: Conclusions are generally balanced but still suggest applicability to training management. Please explicitly state that recommendations are conceptual rather than evidence-based prescriptions.

Authors response: We agree and have revised the Conclusions to clarify that recommendations are conceptual: “Integrating MC and symptom monitoring into routine athlete management may offer a practical approach to optimize performance, support health, and refine injury prevention strategies, although these suggestions remain conceptual and require validation in larger studies.”

Reviewer 3 Report

Comments and Suggestions for Authors

Please view the following as an attempt to aid in the refinement of the manuscript. 

The final sample size (n = 10; UI subgroup n = 4) is extremely small relative to:

• Mixed-model ANOVA with interaction terms
• Canonical correlation analyses with large variable sets

While this is acknowledged indirectly, the risk of overfitting and Type I error remains high.

Recommendations:

• Explicitly acknowledge limited statistical power earlier (Methods or beginning of Results).
• Reframe findings as exploratory or hypothesis-generating, especially for CCA results.
• Consider reducing the number of variables included in CCA or justifying variable selection more explicitly (e.g., theoretical pruning).
• Avoid language that implies causality (e.g., “predicted”) and replace with “associated with” throughout.

 

The CCA section is statistically dense and difficult to interpret, particularly for non-statistical readers. Several canonical functions are reported, but interpretation is cautious even within the Results (“did not consistently align with theoretical expectations”).

Recommendations:

• Justify why CCA was chosen over simpler multivariate approaches given the small sample.
• Consider limiting interpretation to the first canonical function in each analysis, which typically explains the most meaningful shared variance.
• Move some of the more speculative or weak canonical findings to Supplementary Material.
• In the Discussion, clearly distinguish between robust vs. tentative associations.

 

UI is treated both as a categorical grouping variable and as a potential neuromuscular modifier, but its severity range is narrow and may limit interpretability.

Recommendations:

• Emphasize that UI severity was mild to moderate, and conclusions may not generalize to athletes with severe UI.
• Discuss whether a continuous UI score (rather than dichotomization) could be more informative in future work.
• Clarify whether UI status was stable across MC phases.

 

Results are comprehensive but at times overly detailed, especially given the limited number of significant findings.

Recommendations:

• Streamline Table 3 by clearly highlighting statistically relevant outcomes (e.g., visual markers or reduced variable set).
• Reduce repetition between text and tables—allow tables to “carry” more of the data.
• Consider summarizing non-significant findings more concisely.

 

• Minor grammatical issues (e.g., “This Participants characteristics” → “Participant characteristics”).
• Occasional awkward phrasing likely due to translation—recommend professional language editing prior to final submission.
• Ensure consistent spacing (e.g., “UI ,” → “UI,”).

• Use consistent terminology when referring to “ACL injury risk factors” vs. “risk markers.”
• Clarify early that ACL injury itself was not measured, only proxy risk factors.

• Figures are informative but dense.
• Consider simplifying Figure 4 or moving individual-level plots to Supplementary Material.
• Ensure all figures are readable in grayscale (journal formatting consideration).

• More explicitly connect findings to training load management, screening, and return-to-play decisions.
• Strengthen the limitations section by directly addressing:
• Sample size
• Sport heterogeneity
• Multiple testing
• Generalizability

 

Author Response

Please view the following as an attempt to aid in the refinement of the manuscript.

Authors response: We thank the Reviewer for the detailed and constructive feedback, which has greatly contributed to strengthening the manuscript. We have carefully revised the manuscript to address all points raised. We greatly appreciate the time and effort invested in providing these insights.

 

1. The final sample size (n = 10; UI subgroup n = 4) is extremely small relative to:

Mixed-model ANOVA with interaction terms

Canonical correlation analyses with large variable sets

While this is acknowledged indirectly, the risk of overfitting and Type I error remains high.

Recommendations:

Explicitly acknowledge limited statistical power earlier (Methods or beginning of Results).

Reframe findings as exploratory or hypothesis-generating, especially for CCA results.

Consider reducing the number of variables included in CCA or justifying variable selection more explicitly (e.g., theoretical pruning).

Avoid language that implies causality (e.g., “predicted”) and replace with “associated with” throughout.

Authors response: We agree with the reviewer that the final sample size is small, relative to the applied statistical models and that this increases the risk of overfitting and Type I error. To address this, we have explicitly acknowledged the limited statistical power in the methods section and clarified that all analyses were conducted with an exploratory, hypothesis-generating intent.

We have further revised the manuscript to frame the CCA findings as multivariate association patterns rather than confirmatory effects, and we now consistently use non-causal language (e.g., “associated with” instead of “predicted”).

In addition, we have expanded the Methods section to more explicitly justify the selection of variables included in the CCA based on a priori theoretical relevance and the use of summary metrics to limit model complexity. We believe these revisions appropriately contextualize the findings while maintaining their exploratory value.

Finally, to further clarify the exploratory nature of the work and avoid potential misinterpretation, we have added the designation “a pilot study” to the title.

 

2. The CCA section is statistically dense and difficult to interpret, particularly for non-statistical readers. Several canonical functions are reported, but interpretation is cautious even within the Results (“did not consistently align with theoretical expectations”).

Recommendations:

Justify why CCA was chosen over simpler multivariate approaches given the small sample.

Consider limiting interpretation to the first canonical function in each analysis, which typically explains the most meaningful shared variance.

Move some of the more speculative or weak canonical findings to Supplementary Material.

In the Discussion, clearly distinguish between robust vs. tentative associations.

Authors response: According to your suggestion, we have revised the Methods section to more explicitly justify the use of canonical correlation analysis (CCA), emphasizing its suitability for exploring theoretically interdependent variable domains in a hypothesis-generating context, despite the small sample size. Specifically, we added: “Despite the small sample size, CCA was deemed suitable for hypothesis-generating purposes, allowing the identification of shared variance structures across variable sets that may inform future confirmatory studies.”

Regarding the suggestion to move some canonical findings to the Supplementary Material, we considered this carefully. However, we believe it is more accurate and transparent to report all results rather than selectively presenting only a subset. This approach allows readers to fully evaluate the data, while the limitations and tentative nature of some findings are clearly discussed in the Limitations section.

 

3. UI is treated both as a categorical grouping variable and as a potential neuromuscular modifier, but its severity range is narrow and may limit interpretability.

Recommendations:

Emphasize that UI severity was mild to moderate, and conclusions may not generalize to athletes with severe UI.

Discuss whether a continuous UI score (rather than dichotomization) could be more informative in future work.

Clarify whether UI status was stable across MC phases.

Authors response: We thank the reviewer for this important point. We agree and to clarify the scope and interpretability of our findings, we have added the following details:

In the Methods (Section 2.3), we now state: “UI was assessed at a single time point in this study.”

In the Discussion (Section 4.1), we added: “Specifically, three athletes presented with mild UI (ICIQ-UI SF scores 3–4) and one athlete with moderate UI (score 6) [20]. Therefore, the findings may not be generalizable to athletes with more severe UI.”. We also acknowledge the limitations of our assessment (Section 4.1): “Further UI in this study was assessed indirectly using the ICIQ-UI SF questionnaire. Although widely used and validated [21], this instrument does not provide a direct, objective evaluation of pelvic floor muscle function. More detailed and objective evaluations of pelvic floor muscle function, such as palpation or sonographic assessment, could yield a more accurate characterization of pelvic floor muscles [38,39]. It is plausible that more pronounced differences would emerge in athletes with more severe UI or when pelvic floor function is assessed directly, including across multiple MC phases rather than a single time point.”

These revisions emphasize the mild-to-moderate severity of UI in our sample, clarify that UI was assessed at a single time point, and acknowledge that future studies using continuous scores or more objective measures may provide additional insight.

 

4. Results are comprehensive but at times overly detailed, especially given the limited number of significant findings.

Recommendations:

Streamline Table 3 by clearly highlighting statistically relevant outcomes (e.g., visual markers or reduced variable set).

Reduce repetition between text and tables—allow tables to “carry” more of the data.

Consider summarizing non-significant findings more concisely.

Authors response: Thank you for this suggestion. To improve clarity and readability, we have revised Table 2/3 by highlighting statistically significant results using bold text and asterisks. We have also streamlined the accompanying text to reduce repetition, allowing the table to convey more of the data directly (Table 1). Non-significant findings are now summarized more concisely, focusing the reader’s attention on key results while maintaining transparency.

 

5. Minor grammatical issues (e.g., “This Participants characteristics” → “Participant characteristics”).

Authors response: Agreed. All minor grammatical issues, have been corrected throughout the manuscript.

 

6. Occasional awkward phrasing likely due to translation—recommend professional language editing prior to final submission.

Authors response: We agree. The manuscript has been carefully revised for clarity and readability.

 

7. Ensure consistent spacing (e.g., “UI ,” → “UI,”).

Authors response: Agreed. All spacing inconsistencies have been revised throughout the manuscript

 

8. Use consistent terminology when referring to “ACL injury risk factors” vs. “risk markers.”

Clarify early that ACL injury itself was not measured, only proxy risk factors.

Authors response: We thank the reviewer for this suggestion. The manuscript has been carefully proofread and revised to ensure consistent terminology. Additionally, we have clarified early in the Introduction that actual ACL injuries were not measured. Specifically, we added the following sentence to the study aim to improve clarity: “While this study does not measure ACL injuries directly, it focuses on established neuromuscular and biomechanical risk factors associated with increased ACL injury susceptibility.”.

 

9. Figures are informative but dense.

Consider simplifying Figure 4 or moving individual-level plots to Supplementary Material.

Ensure all figures are readable in grayscale (journal formatting consideration).

Authors response: We agree with the reviewer’s comment. To improve clarity and readability of the manuscript, Figure 4 has been moved to the supplemental material (Appendix A.2). Additionally, Figure 1 has been updated to grayscale to ensure readability in journal formatting.

 

10. More explicitly connect findings to training load management, screening, and return-to-play decisions.

Authors response: Thank you for this important suggestion. Accordingly, we have added the following text to Section 4.4 to more explicitly link our findings to training load management, screening, and return-to-play decisions: “Evidence from prior research further underscores the importance of managing symptoms, as unresolved or poorly managed symptoms can impair recovery and sleep, which in turn may influence ACL injury risk factors [47]. While training workload or fatigue alone have not been directly linked to ACL risk factors [48], poor sleep has been associated with detrimental effects on these outcomes [49]. Importantly, the MC itself should not be considered a primary marker of ACL risk. In women’s sports, long-term athletic development and regular assessment of general risk factors and neuromuscular imbalances are essential [50,51]. Systematic screening not only supports injury prevention but also provides benchmarks for return-to-play, enabling safe and effective reintegration into sport. Ongoing monitoring of training adaptation, together with tracking both external and internal loads, facilitates individualized, evidence-based injury-prevention strategies [52].”

 

11. Strengthen the limitations section by directly addressing:

Sample size

       Sport heterogeneity

       Multiple testing

       Generalizability

Authors response: We agree. To address these points, we added the following to the Limitations section: “Given the small sample size, the present findings, especially results of the multivariate analyses, should be considered hypothesis-generating. The identified multivariate association patterns warrant replication in larger, adequately powered cohorts before firm conclusions can be drawn. Additionally, the inclusion of athletes from multiple sports introduces heterogeneity, which may have influenced variability in neuromuscular performance and postural control. Multiple statistical tests were conducted, increasing the risk of Type I error.”

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

I would like to thank the authors for their careful and constructive revisions. The manuscript has improved substantially in clarity, framing, and acknowledgment of methodological constraints. Below, I provide a structured evaluation organized by sections.

Methods: the authors now acknowledge the small final sample (n = 10) and limited generalizability. However, given that 10 out of 24 athletes were excluded due to subtle menstrual dysfunction, the implications for representativeness could be emphasized more explicitly. Consider adding one sentence clarifying that the strict three-step verification enhances internal validity but may reduce ecological validity in elite sport populations.

Statistical analysis: the authors now clearly state that mixed-model ANOVA results are exploratory, CCA findings are hypothesis-generating and limited statistical power increases Type I error risk.These additions improve transparency. However, one important concern remains: The complexity of the canonical correlation analyses relative to n = 10 remains methodologically fragile. Multiple canonical functions (up to four in one model) are reported with high canonical correlations (e.g., rcc = 0.969), which are highly susceptible to instability in small samples. Although the authors acknowledge instability in the limitations, interpretation within the results and discussion still occasionally reads as structured association rather than sample-specific pattern.

Add one explicit sentence stating that canonical coefficients and redundancy indices in such small samples are likely to be sample-specific and may not replicate even in moderately larger cohorts.

Results

Canonical Correlation Analyses: the reporting is technically correct and transparent.Nevertheless, given the small sample and the number of extracted canonical functions, interpretation should remain strictly descriptive. In particular: Statements such as “lower estradiol levels were associated with poorer performance” could be slightly softened to reflect exploratory pattern detection rather than stable biological relationship.

The remaining concerns are primarily related to the interpretative framing of multivariate findings in the context of very small sample size. With minor additional adjustments to reinforce statistical caution, particularly regarding canonical correlation analyses, the manuscript would reach a satisfactory level of methodological coherence for a pilot study.

 

Author Response

I would like to thank the authors for their careful and constructive revisions. The manuscript has improved substantially in clarity, framing, and acknowledgment of methodological constraints. Below, I provide a structured evaluation organized by sections.

Authors response: Thank you for the careful and constructive feedback. All points have been carefully addressed, and we appreciate the detailed engagement, which has helped to improve the manuscript. The changes made in response to your comments are highlighted in the revised version of the manuscript.

 

Methods: the authors now acknowledge the small final sample (n = 10) and limited generalizability. However, given that 10 out of 24 athletes were excluded due to subtle menstrual dysfunction, the implications for representativeness could be emphasized more explicitly. Consider adding one sentence clarifying that the strict three-step verification enhances internal validity but may reduce ecological validity in elite sport populations.

Authors response: Agreed. We have added the following sentence to the limitations section to address this point:  “The strict three-step verification process for determining MC strengthens internal validity but may reduce ecological validity and limit representativeness within elite sport populations, as a substantial proportion of initially screened athletes were excluded due to subtle menstrual dysfunction (10 of 24).“

 

Statistical analysis: the authors now clearly state that mixed-model ANOVA results are exploratory, CCA findings are hypothesis-generating and limited statistical power increases Type I error risk. These additions improve transparency. However, one important concern remains: The complexity of the canonical correlation analyses relative to n = 10 remains methodologically fragile. Multiple canonical functions (up to four in one model) are reported with high canonical correlations (e.g., rcc = 0.969), which are highly susceptible to instability in small samples. Although the authors acknowledge instability in the limitations, interpretation within the results and discussion still occasionally reads as structured association rather than sample-specific pattern.

Authors response:

Thank you for emphasizing this point. Following your suggestion, we added the following clarification to the Methods section: “In addition, canonical correlations are known to be upwardly biased and may reflect over-fitting, and therefore, interpretation focused primarily on the redundancy indices, which provide a more conservative estimate of shared variance between the variable sets and reduce the risk of overstating associations based solely on the magnitude of r_cc.”

We also addressed related concerns regarding cautious interpretation in our later comments.

 

Add one explicit sentence stating that canonical coefficients and redundancy indices in such small samples are likely to be sample-specific and may not replicate even in moderately larger cohorts.

Authors response:

Agreed. Following your suggestion, we have added the following sentence to the Methods section: “Canonical coefficients and redundancy indices should be interpreted with caution, as they are likely to be sample-specific and may not replicate even in moderately larger cohorts.”

 

Results

Canonical Correlation Analyses: the reporting is technically correct and transparent. Nevertheless, given the small sample and the number of extracted canonical functions, interpretation should remain strictly descriptive. In particular: Statements such as “lower estradiol levels were associated with poorer performance” could be slightly softened to reflect exploratory pattern detection rather than stable biological relationship.

Authors response:

Agreed. We carefully revised this section and replaced “were associated” with “might be associated” throughout to better reflect exploratory pattern detection rather than implying stable biological relationships.

 

The remaining concerns are primarily related to the interpretative framing of multivariate findings in the context of very small sample size. With minor additional adjustments to reinforce statistical caution, particularly regarding canonical correlation analyses, the manuscript would reach a satisfactory level of methodological coherence for a pilot study.

Authors response:

We thank the reviewer for these final comments. As noted above, we have carefully addressed all points, including adjustments to the interpretative framing of the multivariate findings and the cautious reporting of canonical correlation analyses in light of our small sample. We hope these revisions meet your expectations and strengthen the manuscript as a pilot study.