Evaluation of Adiponectin as a Metabolic Risk Indicator in the Panamanian Population

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The authors should receive praise for their research which investigates adiponectin levels and their connection to metabolic characteristics in Panamanian adults. The research holds value because Latin America lacks sufficient data about this topic and the researchers should be commended for creating a large study group and conducting ROC analysis for diagnostic evaluation. The research establishes a solid basis for upcoming investigations in this region while delivering important findings for public health initiatives.

The manuscript needs extensive revisions to achieve publication readiness. The study design appears to follow population-based but it actually uses a 3:1 case–control approach with participants from Panama and Panama Oeste who were not selected through probability sampling. The abstract and discussion sections need revision to show the correct study design because the authors should avoid making statements about national representation or screening recommendations. The research team needs to explain their case and control recruitment process by specifying the source of controls and any matching procedures that occurred. The study design as a cross-sectional analysis requires authors to use descriptive language instead of causal statements throughout the entire paper.

The study needs to address its selection criteria. The study population became biased because the authors excluded menopausal women and patients who had COVID-19 or used anti-inflammatory drugs within the past year while adiponectin levels are heavily affected by age and sex. The research should present demographic details about each participant group along with an evaluation of the impact from the exclusion criteria. The study would gain more strength through a sensitivity analysis which includes peri-menopausal women or sample weighting.

The laboratory procedures lack sufficient detail and the measurement units used throughout the study are not uniform. The iFlash 3000 device measured adiponectin through CLIA but the study lacks essential details about calibration procedures and assay precision and quality control measures. The ROC analysis uses µg/mL measurements of adiponectin but Table 1 presents results in ng/mL units. The study requires unit consistency throughout all sections with possible recalculations of summary data if necessary. The researchers need to fix the incomplete results statement regarding sex-specific cutoffs.

The body-composition measurement process receives inadequate presentation in the study. The Tanita BC-545 device generates a proprietary visceral fat rating instead of percentage measurements although the text shows both "visceral fat percentage" and specific thresholds at ≥10. The researchers need to replace the metric with its accurate label while referencing device validation boundaries and should present the results with caution. The primary analysis should use waist circumference as a more dependable marker for adiposity instead of the current metric.

The statistical analysis needs significant improvement to achieve better results. The analysis requires non-parametric methods and median calculations because multiple variables show strong skewness which makes t-tests inappropriate. The research needs to use multivariable regression models to study adiponectin levels while controlling for age and sex and waist circumference and triglycerides and HDL and medication usage. The ROC analysis requires addition of confidence intervals for AUC and sensitivity and specificity values and internal validation through bootstrap methods to detect overfitting. The study needs to evaluate how adiponectin performs when used in combination with basic clinical indicators instead of treating it as an independent diagnostic tool.

The research contains multiple internal errors which need to be fixed. The abstract presents different correlation results than the results section and the case number varies between method descriptions and table data and HDL threshold values remain inconsistent. The research requires a unified code-generated table and figure set that matches all textual information to the presented data.

The analysis of medication effects requires more detailed examination. The discussion presents a hypothesis about hypertension treatment effects but incorrectly classifies thiazolidinediones as antihypertensive drugs when they function as glucose-lowering medications. The authors should use medication use as covariates in their multivariable models to properly interpret the results instead of making speculative statements.

The authors should reduce their statements about diagnostic potential. The AUC value of 0.69 combined with 50% sensitivity and 80% specificity demonstrates limited ability to distinguish between groups. The researchers need to explicitly state that adiponectin shows potential when used with other risk indicators yet it stands alone as an insufficient screening tool.The analysis of figures and tables needs special attention. The research requires complete and readable reference panels with proper units and ROC curves that include reference diagonals and confidence bands. The introduction needs to replace its placeholder text with a properly referenced narrative.

The manuscript requires enhanced reporting standards to achieve better results. The research needs to explain its initial sample size determination and explain data handling methods and show how HbA1c assay standardization was performed. The supplementary materials should contain analysis code and a de-identified data dictionary to enhance research reproducibility. The reference list requires verification to match the journal's formatting requirements.

The study demonstrates that Panamanian adults with metabolic syndrome have lower adiponectin levels and the marker shows restricted diagnostic capabilities. The paper requires substantial modifications to its study design and laboratory reporting and statistical analysis and internal validation and interpretation methods to become a valuable addition to the field. The paper needs extensive revisions before it can reach publication readiness.

Author Response

1. 1. Study Design and Participant Selection

We appreciate the reviewer’s observation regarding the description of the study design and participant selection. The Materials and Methods section has been revised to provide greater clarity on this aspect.

In our study, participants were voluntarily recruited through social media invitations distributed across public Panamanian health and university networks. After confirming eligibility, fasting blood samples were collected at the Immunology Laboratory of the Faculty of Medicine, University of Panama, following standardized protocols for metabolic biomarker evaluation.

The case and control groups were defined according to the diagnostic criteria of the Latin American Diabetes Association (ALAD). Specifically, individuals with central obesity (waist circumference ≥94 cm in men and ≥88 cm in women) plus two or more additional criteria (hypertriglyceridemia, hyperglycemia, low HDL, or elevated blood pressure) were classified as having metabolic syndrome. Controls were participants who did not meet these criteria.

All diagnostic classifications were performed under the supervision of certified endocrinologists, ensuring accurate evaluation and adherence to international clinical standards.

 

2. Inclusion and Exclusion Criteria

We thank the reviewer for this valuable observation regarding the inclusion and exclusion criteria. These aspects have been clarified in the revised manuscript.

Only women undergoing active menopause at the time of sampling were excluded to minimize hormonal variability; however, postmenopausal women are indeed represented in the cohort, as the study included participants up to 60 years of age. This clarification has been added to the text.

Adiponectin is a well-recognized anti-inflammatory adipokine, and our objective was to evaluate its natural circulating levels without pharmacological interference. Therefore, participants who had recently used anti-inflammatory medications were excluded to prevent confounding effects on adiponectin concentration.

Similarly, participants with a recent history of COVID-19 infection were excluded solely to avoid potential post-infectious inflammatory bias, since transient immune activation following SARS-CoV-2 infection may alter adipokine levels.

These clarifications have been incorporated into the Materials and Methods section to ensure transparency in participant selection and to justify the applied exclusion criteria.

 

3. Laboratory Procedures and Data Consistency

We sincerely thank the reviewer for the insightful comments regarding laboratory details and data consistency. The Materials and Methods section has been expanded to include additional information about the analytical procedures performed with the iFlash 3000 (YHLO Biotech).

We now indicate that commercial calibrators and internal quality controls provided by the manufacturer were used in each analytical run to ensure accuracy, precision, and reproducibility of adiponectin quantification. Calibration and validation steps were performed in accordance with the manufacturer’s specifications and the instrument’s standard operating procedures.

All adiponectin concentration units have been standardized to µg/mL throughout the manuscript—including text, tables, and figures—to maintain full consistency with the values used in the ROC analysis. When necessary, numerical values were recalculated to ensure alignment and accuracy.

Additionally, the section describing sex-specific differences in adiponectin concentration has been clarified to include both male and female cutoff values, ensuring that the results are reported completely and transparently.

 

4. Visceral Fat Measurement and Terminology

We appreciate the reviewer’s comment regarding visceral fat terminology. The manuscript now specifies that the Tanita BC-545 provides a visceral fat rating (VFR), a proprietary numerical index rather than a percentage value. We have also included a note describing the validation range reported by the manufacturer and independent studies.

However, we respectfully maintain the inclusion of visceral fat rating as a variable of interest. Recent literature indicates that visceral fat, as estimated by validated bioelectrical impedance devices, shows stronger associations with metabolic risk, insulin resistance, and inflammatory markers than anthropometric measures such as waist circumference. Accordingly, we retained the VFR data and clarified its methodological basis and interpretation in line with current validation studies.

 

5. Statistical Analysis

We thank the reviewer for this valuable comment. The statistical analysis has been substantially improved in the revised manuscript. All variables showing non-normal distributions were reanalyzed using non-parametric tests (Mann–Whitney and Kruskal–Wallis, as appropriate), and the corresponding p-values were corrected accordingly.

To better represent data distribution, all figures have been updated to box-and-whisker plots, displaying the median and interquartile range for each group. These graphical changes enhance the accuracy and transparency of data visualization for adiponectin and related metabolic parameters.

 

6. Figures, Tables, and Internal Consistency

We thank the reviewer for this observation concerning internal consistency between figures, tables, and text. We fully acknowledge the importance of maintaining uniform data presentation throughout the manuscript.

To ensure the highest level of accuracy and visual quality, the authors have decided to request the journal’s professional figure review and editing service, which will be funded by the authors. This service will optimize graphical presentation, verify data alignment, and ensure that all figure labels, correlation values, and thresholds correspond precisely to the text and tables.

All numerical values and dataset identifiers have been rechecked to guarantee full consistency among the abstract, methods, results, and figures.

 

 

Reviewer 2 Report

Comments and Suggestions for Authors

The manuscript meets the publication requirements and is recommended for acceptance.

Author Response

Thank you for your comment.

Reviewer 3 Report

Comments and Suggestions for Authors

This is an interesting article by Lin et al. addressing for the first time the levels of adiponectin and their correlation with several metabolic markers in a Panamanian population.  Although the manuscript is characterized by novelty and a significant sample size, there are some serious methodological issues that, if not addressed, I cannot recommend its publication.

First of all, the sample used in this study is not matched, at least for age and sex. In Table 1 there is no p value for age and it seems like the age is not matched between mets and controls. Also, there is no test presented to prove whether there is a difference between sex distribution in the two groups.

Furthermore, it is not clear in which population the correlation analysis is performed. Is it in the whole sample or just in the mets group? Also, a correlation analysis is not enough, as there are many confounding factors that may affect the relationship between the examined parameters. For example the authors suggest that medication could be responsible for the lack of a significant difference in adiponectin levels between hypertensive and normotensive individuals. In order to avoid assumptions they should perform a regression analysis and adjust for potential covariates, such as age, sex, BMI, medication and any other factor that may affect the examined relationships.

Minor comments:

-there is an extra text (most possible from the template) at the end of your introduction that may have been unintentionally left in the manuscript. Please remove it and revise the aim of the study in a way it is more clear and concise

-please specify whether blood was collected after overnight fast and whether all experiments were performed in replicates?

-in results you cite figure 1 but you mean table 1

Author Response

We sincerely thank the reviewer for their thorough and constructive feedback. We have carefully revised the manuscript to address all the concerns raised.

1. Matching of age and sex:

The study included 310 Panamanian adults, comprising 166 women and 144 men. Among them, 77 participants (43 women and 34 men) were diagnosed with metabolic syndrome, while 233 (132 women and 101 men) served as healthy controls. These distributions have now been clarified and statistically compared in Table 1, confirming no significant sex imbalance between groups

 

 

 

2. Clarification on correlation analysis population and statistical approach

We appreciate this important comment. Correlation analyses were performed using the entire study population (n = 310)to better reflect the continuous metabolic spectrum and overall variability in adiponectin levels. As metabolic syndrome represents a cluster of interrelated metabolic disorders rather than a single condition, participants classified under the ALAD criteria did not necessarily share the same combination of risk factors. Therefore, analyzing the full dataset allowed for a more comprehensive understanding of the relationships between adiponectin and individual metabolic parameters.

In response to another reviewer’s recommendation, all statistical analyses were re-evaluated using non-parametric methods, given the non-normal distribution of several variables and the potential heterogeneity across subgroups.

These methodological refinements provide a more robust and conservative assessment of the associations observed, reducing bias introduced by clinical variability among participants with metabolic syndrome. The revised Materials and Methods and Results sections have been updated accordingly to reflect these changes.

 

 

 

3. Clarification on fasting and experimental replicates:
   We have added a statement in theMaterials and Methodssection confirming that blood samples were collected after an overnight fast of 10–12 hours, and that all biochemical assays, including adiponectin measurement, were performed in duplicate to ensure analytical precision and reproducibility.
4. Text correction and consistency:
   The redundant text mistakenly left at the end of theIntroductionhas been removed, and the study objective has been rewritten to be more concise and aligned with the aims of the research. Additionally, the reference to “Figure 1” in the Results section has been corrected to “Table 1.”

These revisions have strengthened the methodological rigor and clarity of the manuscript. We are grateful to the reviewer for highlighting these key aspects, which have significantly improved the quality and transparency of our work.

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

The revised version of the manuscript shows clear and thoughtful improvements, reflecting that the authors have carefully addressed all previous reviewer comments. The overall structure, presentation, and interpretation have become much stronger, and the study now reads as a coherent and well-executed investigation of adiponectin as a potential biomarker of metabolic risk in the Panamanian population. The paper is well organized and methodologically sound, and the data fully support the conclusions. The language has been refined, figures are appropriately designed, and the discussion is more balanced and evidence-based. This version has substantially increased the scientific and regional value of the work.

Only a few minor adjustments are suggested before the paper proceeds to publication. The abstract could be slightly enhanced by adding the mean adiponectin values for the MetS and control groups, as well as the AUC value (0.69) from the ROC analysis. These quantitative details would make the summary more informative for readers. In the introduction, one or two sentences briefly distinguishing between total and high molecular weight adiponectin would improve clarity and highlight the authors’ analytical focus.

In the methods section, it would be helpful to mention whether data normality was tested prior to using parametric or non-parametric tests and how potential outliers were handled, since adiponectin distribution can be skewed in metabolic populations. In the results, a minor stylistic refinement could involve reducing the repetition between textual descriptions and figure captions—for instance, describing the trends rather than restating all p-values already shown in the figures.

The discussion is now clear and well-balanced. However, the authors might consider adding one short comparative statement referencing how their reported adiponectin cutoff (6.9 µg/mL) aligns with or differs from values reported in similar Latin American or international cohorts. This addition would provide a broader context and strengthen the relevance of their findings beyond the national level.

Overall, this is an excellent revision that meaningfully contributes to understanding metabolic risk biomarkers in a Latin American population. With these minor editorial refinements for conciseness, consistency, and contextual comparison, the manuscript will be fully ready for publication in Obesity.

Author Response

We appreciate the reviewer’s constructive suggestions. In response, the abstract has been updated to include the mean adiponectin values for both the MetS and control groups, as well as the AUC value from the ROC analysis. The introduction was expanded to include additional information distinguishing total and high-molecular-weight (HMW) adiponectin, emphasizing their physiological relevance.

Moreover, comparative data from other international and Latin American cohorts have been incorporated into the discussion to contextualize the reported cutoff value. However, it is important to note that there are still very few studies available in the region, which limits the extent of direct regional comparison.

These modifications strengthen the clarity, completeness, and contextual relevance of the manuscript as suggested.

Reviewer 3 Report

Comments and Suggestions for Authors

The authors did not succesfully addressed my comments. They say that "age and sex distributions have now been clarified and statistically compared in Table 1, confirming no significant sex imbalance between groups" but there is no such analysis presented in Table 1 (at least at the version I have access to). Also, they have not addressed my comment on adjustment for potential covariates, which is very important in such a study.

Author Response

We thank the reviewer for this valuable observation. In response, a new supplementary table has been created including the distribution by sex and the main clinical variables of participants with and without metabolic syndrome. This addition provides the statistical comparison that was previously missing.

Furthermore, following the reviewer’s recommendation, a multivariable linear regression analysis was performed to adjust for potential covariates in relation to the studied variables. These results are now presented in Supplementary Table 2.

These updates directly address the reviewer’s comment and improve the statistical rigor and interpretability of the findings.