Long-Term Resistance Training Induces Structural Muscle Changes: A Comparative Study

Fig. 1. Illustration of skeletal muscle structure and obtained example images of ACSAmax of the biceps brachii (A), muscle fiber (B), myofibril (C), and myofilament (D). The arrows in D show how the myosin spacing (distance) was measured. The illustration of skeletal muscle structure was purchased from Dreamstime.com (file ID: 80735107 created by Legger) and adapted with permission.

• Overview: Figure 1 presents a multi-scale visualization of skeletal muscle structure, ranging from the whole muscle to the myofilaments. It combines a schematic illustration with actual microscope images.
• Panel Breakdown: Panel A: MRI image of biceps brachii showing the maximal anatomical cross-sectional area (ACSAmax). Panel B: Light microscopy image of muscle fibers (Type I and II). Panel C: Transmission electron microscopy (TEM) image of a myofibril. Panel D: Higher magnification TEM image of a sarcomere showing myofilaments and arrows indicating myosin spacing measurement.
• Source and Adaptation: The caption clearly states that the muscle structure illustration was sourced from Dreamstime.com and adapted with permission, providing appropriate attribution.

• Image Quality and Resolution: While the images serve the purpose of illustration, the resolution, particularly in panels B, C, and D, could be higher for better clarity and detail. Higher resolution images would allow for clearer visualization of individual structures and enhance the scientific value.
• Methodological Transparency: The caption mentions MRI and TEM but lacks details on sample preparation, staining (for panel B), and image processing. Providing these details in the methods section or as supplementary information would enhance transparency and reproducibility.
• Measurement Precision: The method for measuring myosin spacing is visually indicated, but the precision of the measurement is not specified. Including information on measurement error or variability would strengthen the scientific rigor.
• Scale and Context: While different magnifications are used to show different structural levels, the lack of scale bars in panels B, C, and D makes it difficult to judge the actual size of the structures. Adding scale bars would provide essential context.

• Clarity and Organization: The figure is well-organized and the caption clearly labels each panel. The multi-scale approach effectively conveys the hierarchical nature of muscle structure.
• Integration with Text: The text reference to "several hundred myofilaments" directly relates to panel D. However, the text could benefit from more explicit connections to the other panels to guide the reader through the figure.
• Accessibility: The figure is accessible to a scientific audience. For a broader audience, adding labels of key structures within each image (e.g., muscle fiber, fascicle, myofibril, sarcomere, actin, myosin) would improve understanding. Explaining ACSAmax in simpler terms would also broaden accessibility.
• Visual Appeal: The figure could be visually improved by using consistent image quality and resolution across all panels. Adding a title to the figure itself (e.g., "Hierarchical Structure of Skeletal Muscle") would enhance its standalone readability.

Fig. 2. ACSAmax of LRT and UNT males. * denotes a significant difference between groups (P<0.05) by a t-test. Individual plots together with box (lower-upper quartile) and whisker (SD) plots are shown. The cross mark and line in the box indicate the mean and median, respectively.

• Data Presentation: Figure 2 compares the maximal anatomical cross-sectional area (ACSAmax) of the biceps brachii muscle between long-term resistance-trained (LRT) and untrained (UNT) males. It uses a combined approach of showing individual data points, box plots, and whiskers to represent the data distribution for each group.
• Box Plot Components: The boxes in the box plot represent the interquartile range (IQR), spanning from the 25th percentile (bottom of the box) to the 75th percentile (top of the box). The horizontal line within each box marks the median value. Whiskers extend from the boxes, and the caption states they represent the standard deviation (SD). Individual data points outside the whiskers are plotted separately.
• Statistical Significance: The asterisk (*) indicates a statistically significant difference (P < 0.05) between the LRT and UNT groups based on a t-test.

• Choice of Statistical Test: While a t-test is commonly used to compare means between two groups, its validity depends on meeting certain assumptions, namely normal distribution and equal variances within each group. The caption doesn't mention whether these assumptions were verified. If the assumptions aren't met, a non-parametric test like the Mann-Whitney U test would be more appropriate. The visual presentation of the data in the figure hints at a possible deviation from normality, especially for the LRT group.
• Effect Size: The inclusion of Cohen's d (d = 3.07) is commendable, as it quantifies the magnitude of the difference between groups. Classifying the effect size as "large" aligns with standard interpretations of Cohen's d values.
• Sample Size: Although individual data points are shown, the figure caption omits the exact sample size (n) for each group. Explicitly stating 'n' in the caption would improve transparency and interpretation.
• Whisker Representation: The caption indicates that whiskers represent standard deviation (SD). This is unconventional for box plots, where whiskers typically represent 1.5 times the IQR or the range of data within a specified distance from the IQR. This discrepancy needs clarification as it can lead to misinterpretation of the data spread. The authors should clarify what the whiskers represent or use standard box plot conventions.

• Clarity of Message: The figure effectively conveys the primary message of a significant difference in ACSAmax between LRT and UNT groups. The box plots facilitate visual comparison of central tendency and data spread.
• Text Integration: The accompanying text clearly refers to the figure and presents the key statistical findings (means, p-value, effect size). The text and figure complement each other effectively.
• Accessibility for Wider Audience: While clear for a scientific audience, the figure's accessibility could be improved for a broader audience by explaining ACSAmax more clearly in the caption (e.g., "ACSAmax, a measure of muscle size...").
• Visual Presentation: The visual clarity could be enhanced by making the individual data points more distinct (e.g., using different shapes or colors). Adding a clear y-axis label "ACSAmax (cm²)" would also improve readability and interpretation.

Fig. 3. Muscle fiber type composition by number (A) and by area (B), and type I fiber area (C) and type II fiber area (D) of LRT and UNT males. * denotes a significant difference between groups (P<0.05). To compare the groups, Mann-Whitney U test (C) and t-tests (A, B, D) were used. Individual plots together with box (lower-upper quartile) and whisker (SD) plots are shown. The cross mark and line in the box indicate the mean and median, respectively.

• Fiber Type Composition (A & B): Panels A and B present the relative proportions of Type I and Type II muscle fibers in LRT (long-term resistance-trained) and UNT (untrained) males. Panel A shows the composition by the number of fibers, while Panel B shows the composition by the area occupied by each fiber type.
• Fiber Area (C & D): Panels C and D illustrate the cross-sectional area of individual muscle fibers. Panel C focuses on Type I fibers, and Panel D focuses on Type II fibers, comparing the fiber areas between the LRT and UNT groups.
• Data Visualization: All panels utilize box plots with individual data points overlaid. The box plots display the median, interquartile range (IQR), and potential outliers. Whiskers, according to the caption, represent standard deviation (SD). Cross marks within boxes indicate the mean.

• Statistical Analysis: The use of t-tests for panels A, B, and D is appropriate for comparing means between two independent groups, assuming the data meet the assumptions of normality and equal variances. The caption does not confirm whether these assumptions were met. The Mann-Whitney U test for panel C is suitable for non-normally distributed data. The specific tests used for each panel are clearly stated, which is good practice. However, providing the rationale for choosing the Mann-Whitney U test for panel C would further strengthen the methodology.
• Effect Size: Reporting Cohen's d values provides valuable information about the magnitude of the observed effects. The characterization of the effect sizes as "small" and "trivial" is consistent with conventional interpretations of Cohen's d.
• Data Presentation: Whiskers: The caption states that whiskers represent standard deviation (SD), which is not typical for box plots. Whiskers usually represent 1.5 times the IQR or the data range within a certain distance from the IQR. This deviation from standard practice should be clarified or corrected to avoid misinterpretations of data spread.
• Missing Sample Sizes: The caption omits the sample sizes (n) for each group in each panel. Including this information directly in the caption would enhance transparency and allow readers to better assess the statistical power.

• Clarity and Message: The figure effectively communicates the main finding of no significant difference in fiber type composition between LRT and UNT groups. The box plots facilitate easy visual comparison of distributions.
• Integration with Text: The reference text explicitly links to the figure panels and presents the key statistical results, creating a strong connection between the visual and textual information.
• Accessibility: While generally understandable for a scientific audience, the figure's accessibility could be improved for a broader audience by providing clearer explanations of terms like "fiber type composition" and differentiating between "composition by number" and "composition by area" within the caption itself.
• Visual Presentation: The figure's visual organization could be improved by separating panels A/B and C/D more distinctly. Adding axis labels (e.g., "% of Fibers", "Fiber Area (μm²)") and more descriptive panel titles (e.g., "Fiber Type Composition by Number", "Type I Fiber Cross-Sectional Area") would enhance clarity. Using distinct colors or symbols for individual data points would further improve readability.

Fig. 4. Average fiber area across type I and II fibers (A) and estimated fiber number (B), and relationships of ACSAmax with average fiber area (C) and fiber number (D) of LRT and UNT males. * denotes a significant difference between groups (P<0.05) by a t-test. Individual plots together with box (lower-upper quartile) and whisker (SD) plots are shown. The cross mark and line in the box indicate the mean and median, respectively.

• Panel A: Average Fiber Area: This panel compares the average cross-sectional area of muscle fibers (combining both Type I and Type II fibers) between LRT and UNT groups using box plots and individual data points.
• Panel B: Estimated Fiber Number: This panel compares the estimated total number of muscle fibers within the measured ACSAmax between LRT and UNT groups, again using box plots and individual data points. It's important to note that this is an *estimated* fiber number, as explained in the methods section of the paper.
• Panels C & D: Correlations with ACSAmax: These panels present scatter plots showing the relationship between ACSAmax and average fiber area (Panel C) and between ACSAmax and estimated fiber number (Panel D). The correlation coefficients (r) and p-values are displayed on each plot.
• Data Visualization: All panels use box plots to represent the interquartile range (IQR), median, and potential outliers. Individual data points are overlaid on the box plots. Whiskers, according to the caption, represent standard deviation (SD). The mean is indicated by a cross mark within each box.

• Statistical Analysis (A & B): The use of a t-test is appropriate for comparing means between two independent groups (LRT and UNT) in panels A and B, provided the assumptions of normality and equal variances are met. The caption does not state whether these assumptions were checked. If the assumptions are violated, a non-parametric test would be more suitable.
• Estimated Fiber Number (B): The method for estimating fiber number (dividing ACSAmax by average fiber area) has inherent limitations, as acknowledged in the paper. It's crucial to interpret these results with caution, recognizing that they are estimations, not direct measurements.
• Correlation Analysis (C & D): Presenting correlations between ACSAmax and fiber area/number is relevant. However, the type of correlation (Pearson, Spearman) is not specified in the caption. Additionally, potential confounding factors influencing these relationships should be considered and discussed.
• Whisker Representation: As in previous figures, the caption indicates that whiskers represent standard deviation (SD), which is not standard for box plots. This needs clarification or correction, as it can mislead readers about the data spread.

• Clarity and Message: The figure clearly communicates the significant difference in average fiber area and estimated fiber number between LRT and UNT groups. The scatter plots effectively visualize the relationships with ACSAmax.
• Integration with Text: The reference text directly refers to panel A and provides key statistical results. However, the text could be strengthened by explicitly mentioning the findings related to fiber number and the correlations presented in panels C and D.
• Accessibility: The figure is generally accessible to a scientific audience. For a wider audience, explaining the concept of "estimated fiber number" and the implications of the correlations in simpler terms would be beneficial.
• Visual Presentation: The visual clarity could be improved by adding axis labels to all panels (e.g., "Average Fiber Area (μm²)", "Estimated Fiber Number", "ACSAmax (cm²)"). Using different symbols or colors for LRT and UNT data points in panels C and D would enhance readability. Clearly distinguishing visually between panels A/B and C/D would improve the overall organization.

Fig. 5. Myofibril area (A), myofibril number per fiber (B), and total myofibril number (C), and relationships of ACSAmax with myofibril area (D), myofibril number per fiber (E), and total myofibril number (F) of LRT and UNT males. * denotes a significant difference between groups (P<0.05) by a t-test, and # denotes a tendency toward significance (P = 0.072) by Mann-Whitney U test. Individual plots together with box (lower-upper quartile) and whisker (SD) plots are shown. The cross mark and line in the box indicate the mean and median, respectively.

• Panels A, B, and C: Myofibril Characteristics: These panels compare myofibril area (A), myofibril number per fiber (B), and total myofibril number (C) between LRT and UNT groups using box plots with overlaid individual data points. Panel C represents the total number of myofibrils within the entire muscle cross-section (ACSAmax).
• Panels D, E, and F: Correlations with ACSAmax: These panels display scatter plots illustrating the relationships between ACSAmax and myofibril area (D), myofibril number per fiber (E), and total myofibril number (F). Correlation coefficients (r) and p-values are included in each panel.
• Statistical Significance and Tendency: An asterisk (*) denotes a statistically significant difference (P < 0.05) between groups. A hashtag (#) indicates a tendency towards significance (P = 0.072).
• Data Visualization: Box plots are used throughout, showing the median, quartiles, and potential outliers. Whiskers, as stated in the caption, represent standard deviation (SD). Individual data points are plotted, and the mean is marked by a cross within each box.

• Statistical Tests: The use of t-tests for significant differences is appropriate for comparing means between two independent groups, assuming normality and equal variances. The Mann-Whitney U test is suitable for non-normally distributed data. However, the caption doesn't explicitly state whether the assumptions of the t-test were verified. The rationale for using the Mann-Whitney U test for myofibril area (Panel A) should be clarified, given that a t-test is typically used for comparing means.
• Correlation Analysis: While exploring the relationships between ACSAmax and myofibril characteristics is relevant, the caption doesn't specify the type of correlation used (Pearson or Spearman). Furthermore, potential confounding variables should be considered and discussed in the paper.
• Whisker Representation: The caption's indication that whiskers represent standard deviation (SD) is inconsistent with standard box plot conventions. Whiskers typically represent 1.5*IQR or the data range within a specified interval from the quartiles. This needs to be corrected to avoid misinterpretations.
• Data Interpretation (Panel A): Describing the -16% difference in myofibril area as a "tendency" is appropriate given the p-value of 0.072, which is slightly above the conventional significance threshold of 0.05.

• Clarity and Message: The figure effectively communicates the main findings related to myofibril characteristics and their relationships with ACSAmax. The use of scatter plots is appropriate for visualizing correlations.
• Integration with Text: The reference text directly links to panels A and D and summarizes the key findings. However, the text could be more comprehensive by mentioning the significant findings in panels B, C, E, and F.
• Accessibility: For a broader audience, explaining terms like "myofibril area", "myofibril number per fiber", and "total myofibril number" in simpler language within the caption would improve understanding.
• Visual Presentation: Adding axis labels to all panels (e.g., "Myofibril Area (μm²)", "Myofibril Number per Fiber", "Total Myofibril Number", "ACSAmax (cm²)") is essential for clear interpretation. Using different colors or symbols to distinguish LRT and UNT data points in the scatter plots would improve readability. Visually separating the panels into two distinct groups (A/B/C and D/E/F) would enhance the figure's organization.

Fig. 6. Myosin interspace (A) and its relationship with ACSAmax (B) of LRT and UNT males. * denotes a significant difference between groups (P<0.05) by Mann-Whitney U test. Individual plots together with box (lower-upper quartile) and whisker (SD) plots are shown. The cross mark and line in the box indicate the mean and median, respectively.

• Panel A: Myosin Interspace: This panel compares the myosin interspace (a measure of the distance between myosin filaments) between LRT and UNT groups using box plots and individual data points.
• Panel B: Correlation with ACSAmax: This panel presents a scatter plot showing the relationship between myosin interspace and ACSAmax. The correlation coefficient (r) and p-value are displayed on the plot.
• Data Visualization: Both panels utilize box plots to represent the median, interquartile range (IQR), and potential outliers. Individual data points are overlaid, and whiskers, according to the caption, represent standard deviation (SD). The mean is indicated by a cross mark within each box.

• Statistical Test (Panel A): The use of the Mann-Whitney U test is appropriate for comparing groups if the data are not normally distributed. However, the justification for choosing this test over a t-test is not explicitly stated in the caption or the referenced text. Providing this rationale would strengthen the methodology.
• Correlation Analysis (Panel B): Presenting the correlation between myosin interspace and ACSAmax is relevant. However, the type of correlation (Pearson or Spearman) is not specified. Additionally, it's important to consider and discuss potential confounding factors that might influence this relationship.
• Whisker Representation: The caption indicates that whiskers represent standard deviation (SD), which is not standard practice for box plots. Whiskers typically represent 1.5 times the IQR or the data range within a specified interval from the IQR. This needs correction to avoid misinterpretation of data spread.
• Measurement Precision: The methods section should ideally provide details about the precision and reliability of the myosin interspace measurements. This information is crucial for assessing the validity and robustness of the findings.

• Clarity and Message: The figure effectively communicates the main finding of a smaller myosin interspace in LRT compared to UNT and its negative correlation with ACSAmax. The use of a scatter plot is appropriate for visualizing the correlation.
• Integration with Text: The reference text clearly refers to both panels of the figure and summarizes the key findings. The text and figure complement each other well.
• Accessibility: For a broader audience, explaining "myosin interspace" in simpler terms within the caption would improve understanding. For example, defining it as the distance between myosin filaments and explaining its relevance to muscle contraction would be helpful.
• Visual Presentation: Adding axis labels to both panels (e.g., "Myosin Interspace (nm)", "ACSAmax (cm²)") is crucial for proper interpretation. The visual presentation could be further improved by using different symbols or colors for the individual data points in the scatter plot (Panel B) to enhance readability.

• Content: Table 1 presents descriptive statistics for various physical characteristics of the participants in the LRT (long-term resistance-trained) and UNT (untrained) groups. These characteristics include age, height, humerus length, body mass, and IPAQ (International Physical Activity Questionnaire) value.
• Organization: The table is organized into columns representing the variable measured, the mean value for the LRT group, the mean value for the UNT group, the p-value for the comparison between groups, and Cohen's d (effect size).
• Statistical Significance: Asterisks (*) are used to indicate statistically significant differences (P < 0.05) between the LRT and UNT groups.
• Effect Size Interpretation: The table includes Cohen's d values and labels them with their corresponding interpretations (large, medium, small) to provide context for the magnitude of the differences between groups.

• Data Reporting: Reporting means and standard deviations (indicated by ±) is standard practice for descriptive statistics. Including p-values and effect sizes (Cohen's d) strengthens the analysis by providing measures of statistical significance and practical significance, respectively.
• Missing Sample Sizes: While the table caption mentions LRT and UNT individuals, it doesn't explicitly state the sample size (n) for each group. Including 'n' in the table or caption would improve transparency and allow readers to better assess the statistical power.
• IPAQ Value: The inclusion of IPAQ values is valuable for characterizing the participants' physical activity levels. However, the table could benefit from specifying the version of the IPAQ used (short or long form) and providing a brief explanation of how the IPAQ score is interpreted.
• Justification of Statistical Tests: The table doesn't specify which statistical tests were used to compare the groups for each variable. While it's likely that t-tests were used for most variables, explicitly stating the tests used would enhance transparency. Given the inclusion of Cohen's d, which is typically associated with t-tests, it would be beneficial to confirm this or specify if other tests were used for any variables.

• Clarity and Organization: The table is generally well-organized and easy to read. The inclusion of clear column headers and units makes the data readily interpretable.
• Integration with Text: The reference text clearly directs the reader to Table 1 for descriptive statistics, creating a good connection between the text and the table.
• Accessibility: The table is generally accessible to a scientific audience. For a broader audience, providing brief explanations of less common terms like "IPAQ" and "humerus length" in a footnote would improve understanding. Defining the abbreviations LRT and UNT in the table caption itself would also be helpful.
• Visual Presentation: The visual presentation of the table is functional but could be improved. Using clearer formatting (e.g., bolding the variable names, adding lines to separate rows and columns) would enhance readability. Consider aligning the numerical data by decimal point for easier comparison.

Fig. 7. Summary of the differences in whole muscle, muscle fiber and ultrastructural variables of LRT compared with UNT males. Data are percentage differences in group mean values.

• Data Presentation: Figure 7 provides a summarized visual representation of the key findings, showing the percentage differences in various muscle parameters between LRT (long-term resistance-trained) and UNT (untrained) males. It uses a bar graph format to display these differences.
• Variables Compared: The figure compares the following variables: ACSAmax (whole muscle cross-sectional area), muscle fiber area, muscle fiber number, myofibril area, myofibril number per fiber, total myofibril number, and myosin spacing.
• Percentage Differences: The bars represent the percentage change in LRT values compared to UNT values. Positive values indicate an increase in LRT, while negative values would indicate a decrease.
• Statistical Significance: Asterisks (*) indicate statistically significant differences (P < 0.05). A hashtag (#) denotes a tendency toward significance (P = 0.07).

• Summary of Findings: The figure serves as a clear visual summary of the main findings presented in the results section. It effectively highlights the key differences between the LRT and UNT groups.
• Data Accuracy: The accuracy of the percentage differences presented in the figure relies on the accuracy of the underlying data and calculations presented in the results section. It's assumed that these values are correctly calculated and represented.
• Missing Error Bars: While the figure shows percentage differences, it lacks error bars or confidence intervals. Including error bars would provide a better representation of the variability and statistical significance of the observed differences.
• Choice of Percentage Change: Presenting percentage change can be misleading if the baseline values (UNT group) are very small. It's important to consider the absolute values as well when interpreting percentage differences.

• Clarity and Conciseness: The figure effectively summarizes the main findings in a clear and concise visual format. The bar graph allows for easy comparison of the magnitude and direction of changes between groups.
• Integration with Text: The reference text in the conclusion directly refers to Figure 7, creating a clear link between the text and the visual summary.
• Accessibility: The figure is generally accessible to a scientific audience. For a broader audience, providing brief explanations of the variables (e.g., ACSAmax: muscle size, myosin spacing: distance between muscle filaments) within the figure or caption would enhance understanding.
• Visual Presentation: The visual presentation could be improved by adding a clear title to the figure (e.g., "Summary of Muscle Adaptations to Long-Term Resistance Training") and more descriptive x-axis labels. Using different colors for the bars representing significant changes versus tendencies would further enhance the visual impact and clarity.