Sixty Years After a Coal Mine Disaster: Serum Metabolomic Profiles in Older Adults with Long-Term Sequelae of Carbon Monoxide Poisoning: A Cross-Sectional Study

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The authors conducted a pilot case-control study of survivors of the 1963 coal mine disaster to examine the long-term effects of carbon monoxide (CO) poisoning. Compared to the control group, survivors exhibited an altered serum metabolite profile, characterized by elevated levels of amino acid–related metabolites and decreased levels of ketone-body and purine-related metabolites. Additionally, a reduction in circulating BDNF levels was observed. Although the sample size is small and the magnitude of change is moderate, the authors conclude that CO poisoning leads to persistent systemic metabolic remodeling in late-life CO sequelae. Consistent with previous findings, this metabolic pattern is accompanied by enduring cognitive and functional impairment

This is an interesting and important study that is relevant in this area. The statements and conclusions are consistent with the evidence drawn coherent and supported by the citations. The study design, data analysis, statistical analysis., the references, data availability and ethics statements are appropriate and adequate. This article can be accepted for publication after revision.

Major

1. Please provide a brief description of the 1963 CO exposure. Clarify the nature of the event, its duration, and the immediate clinical consequences observed at that time.
2. The Discussion appears disjointed and difficult to read. The authors are requested to structure it more clearly, perhaps use subheadings and reduce the overall length.

Minor

1. Make the text in Fig. 1 larger and clearer.

Author Response

1. Responses to Reviewer #1

Comment 1 (Major): Add a brief description of the 1963 CO exposure

Reviewer comment: Please provide a brief description of the 1963 CO exposure. Clarify the nature of the event, its duration, and the immediate clinical consequences observed at that time.

Response: Thank you for this important comment. We added one sentence to the Introduction to briefly strengthen background information regarding the nature of the disaster (underground coal-dust explosions and subsequent CO spread due to continued ventilation), the delay until rescue, and the immediate clinical consequences (fatal and non-fatal cases, and need for hospitalization). In addition, we noted that acute-phase severity data (duration of impaired consciousness and whether acute care was inpatient or outpatient) are available from long-term records, citing our prior cohort report [9] (Table 1 therein) and adding this information near Methods/Participants.

Location of changes: Introduction (Background) / Methods (around Participants).

Revised text (excerpt): During an afternoon shift change, runaway rail cars triggered coal-dust explosions deep underground; because the ventilation fan continued to operate, CO spread throughout the mine and trapped nearly 1,400 workers far from the entrance. (Lines 48–52)

Revised text (excerpt): [See revised manuscript, Methods/Participants section, describing recruitment, groups, eligibility, exclusions, and availability of acute-phase severity data; Lines 99–102].

 

Comment 2 (Major): Structure and shorten the Discussion

Reviewer comment: The Discussion appears disjointed and difficult to read. The authors are requested to structure it more clearly, perhaps use subheadings and reduce the overall length.

Response: Thank you. We restructured the Discussion using subheadings (Sections 4.1–4.6) to clarify the logical flow (key findings → mechanistic interpretation → BDNF → exploratory correlations → implications/future directions). We also removed redundant examples and comparisons and shortened multiple passages while preserving the overall rationale.

Location of changes: Discussion (Sections 4.1–4.6).

 

Comment 3 (Minor): Enlarge the text in Figure 1

Reviewer comment: Make the text in Fig. 1 larger and clearer.

Response: We enlarged the font size in Figure 1 and simplified group-label wording to improve readability. We also replaced Figure 1 with a corrected version after fixing minor errors (numbers and group labels).

Location of changes: Figure 1 (replaced).

Conclusion

We believe that these revisions address the reviewers’ concerns (clarification of the statistical framework and exploratory interpretation, potential confounding, positioning of correlation analyses, platform-dependent interpretation, more cautious biomarker language, and improved reader accessibility) with minimal additional text while improving clarity. We greatly appreciate your continued guidance.

Author Response File: Author Response.docx

Reviewer 2 Report

Comments and Suggestions for Authors

This manuscript presents a rare and scientifically compelling investigation of long-term survivors of carbon monoxide (CO) poisoning nearly 60 years after the Miike–Mikawa coal mine disaster. The integration of targeted serum metabolomics with detailed cognitive, functional, and behavioral assessments addresses an important gap in the literature regarding the persistence of systemic metabolic alterations decades after toxic hypoxic brain injury. The study is novel in both its cohort and temporal scope, and the authors demonstrate commendable rigor in participant characterization, sensitivity analyses (including propensity-score matching), and cautious interpretation of exploratory findings.

 With several targeted revisions to strengthen methodological clarity, statistical framing, and interpretability, this work has strong potential to make a meaningful contribution to the field. Here is the list of the concerns/suggestions:

 

• While BH-FDR q-values are reported in supplementary materials, the main text and tables emphasize nominal p-values (p < 0.05), which may overstate significance given the number of metabolites tested. I recommend stating in the main Results and Discussion that no metabolites survived FDR correction (if applicable), and clarify that findings are exploratory. Also, please consider highlighting effect sizes and consistency across unmatched/matched analyses as the primary evidence rather than statistical significance alone.

 

• Amino acids (e.g., valine, alanine, betaine) and ketone bodies are strongly influenced by diet and activity, yet these variables were not assessed. It would be better to expand the Limitations section to more specifically acknowledge the absence of dietary and physical activity data. Also, in the Discussion, clarify that while consistent metabolite patterns may reflect disease-related remodeling, nutritional factors cannot be excluded.

 

• Several correlations between metabolites and functional outcomes are reported, but most disappear after group adjustment. The authors could emphasize more clearly (earlier in the Results or Discussion) that these correlations are largely driven by between-group differences rather than within-group variability. The authors may consider reframing these analyses as descriptive pattern exploration rather than associative evidence.

 

• The panel includes 1,183 annotated metabolites, but only 245 were detected, and pathway coverage is uneven. It could be better if the authors briefly discuss how the characteristics of the CE–MS targeted panel may bias detection toward polar metabolites (e.g., amino acids, nucleotides). Also, please clarify that absence of lipid or acylcarnitine differences does not exclude mitochondrial dysfunction but may reflect analytical scope.

 

• The manuscript suggests potential biomarker utility of the metabolite signature and BDNF. Further temper this language by clearly distinguishing hypothesis-generating signatures from validated biomarkers. Please explicitly state what type of future study (e.g., longitudinal tracking, intervention response, external validation cohort) would be required to support biomarker development.

 

• There are a few minor concerns:

1. Terminology consistency: Please ensure consistent use of “3-hydroxybutyric acid” vs. “β-hydroxybutyrate” throughout the manuscript.
2. Figure and table clarity: I suggest to consider adding a schematic or pathway-based summary figure integrating amino acid, ketone, purine metabolism, and BDNF for reader accessibility.
3. Grammar and formatting: Minor typographical inconsistencies (e.g., spacing, symbols in tables) should be corrected during copyediting.

Author Response

2. Responses to Reviewer #2

Comment 1 (Major): Clarify BH-FDR and the exploratory nature of findings

Reviewer comment: While BH-FDR q-values are reported in supplementary materials, the main text and tables emphasize nominal p-values (p < 0.05) ... (if applicable) state clearly that no metabolites survived BH-FDR correction ... and clarify that findings are exploratory.

Response: Thank you. We revised the Results framework to avoid over-emphasizing nominal p-values by: (i) presenting effect sizes (Hedges’ g) and 95% CIs in Table 2; (ii) reporting BH-FDR q-values in Supplementary Table S5; and (iii) emphasizing direction concordance before and after matching as a sensitivity perspective. Because no metabolite survived BH-FDR correction (q < 0.05) in Supplementary Table S5, we explicitly stated this in the Results and clarified that the findings are exploratory and hypothesis-generating. We also aligned the Statistical Analysis section and Table 2 footnote accordingly. We additionally standardized the footnotes in Supplementary Tables S2–S5 to match the main tables (footnote edits only).

Location of changes: Results (Table 2 description) / Methods (Statistical Analysis) / Table 2 footnote / Supplementary Table S5 (e.g., Results: Lines 266–268; Table 2 footnote).

Revised text (excerpt): No metabolite survived BH-FDR correction (q < 0.05) across metabolites with available p values (Supplementary Table S5); therefore, these findings should be interpreted as exploratory and hypothesis-generating. (Lines 266–268)

 

Comment 2 (Major): Specify unmeasured diet and physical activity as limitations

Reviewer comment: Amino acids ... and ketone bodies are strongly influenced by diet and activity ... expand the Limitations ... clarify ... nutritional factors cannot be excluded.

Response: Thank you. In the Discussion (amino acids/betaine paragraph), we added a statement that nutritional factors and physical activity may influence these metabolites and cannot be excluded. In the Limitations, we also specified that we did not collect quantitative dietary intake (e.g., recent protein intake or supplementation) or objective physical-activity measures, clarifying potential residual confounding.

Location of changes: Discussion (Section 4.2) / Limitations.

Revised text (excerpt): In particular, we did not collect quantitative dietary intake (e.g., recent protein intake, supplementation) or objective physical-activity measures. Revised text (excerpt): Because circulating amino acids, betaine, and ketone bodies are sensitive to nutrition and activity, residual confounding by these factors remains possible. (Lines 491–494)

 

Comment 3 (Major): Emphasize correlations may be driven by between-group differences

Reviewer comment: Several correlations ... disappear after group adjustment ... emphasize earlier that correlations may be driven by between-group differences ...

Response: Thank you. We added an early statement in the main text introducing Table 3 that correlations can be driven by between-group separation. We also reported group-adjusted partial correlations in Supplementary Table S2. In addition, we removed the statement implying p < 0.05 as “significant” from the Table 3 footnote and clarified that p-values are nominal/descriptive.

Location of changes: Results (Table 3 introduction) / Table 3 footnote / Supplementary Table S2.

Revised text (excerpt): Because correlations can be driven by between-group separation, group-adjusted partial correlations are reported in Supplementary Table S2. (Lines 288–291)

 

Comment 4 (Major): CE–MS panel bias and interpretation of lipid-null findings

Reviewer comment: Only a subset of metabolites were detected ... CE–MS may be biased toward polar metabolites ... absence of lipid/acylcarnitine differences does not exclude mitochondrial dysfunction.

Response: Thank you. In Methods (Section 2.6), we added that CE–MS–based targeted panels preferentially capture polar/ionic metabolites and therefore provide uneven pathway coverage (e.g., relatively limited lipid coverage). Accordingly, null findings for lipid-related analytes should be interpreted within the analytical scope of this platform. In the Discussion, we clarified that lipid-null findings do not exclude mitochondrial dysfunction, while suggesting that the observed long-term CO-associated signature in this dataset is more strongly reflected in amino-acid, ketone-body, and purine/energy-related pathways.

Location of changes: Methods (Section 2.6) / Discussion (Section 4.3).

Revised text (excerpt): Because CE–MS-based panels preferentially capture polar/ionic metabolites, pathway coverage is uneven (e.g., relatively limited lipid coverage); therefore, null findings for lipid-related analytes should be interpreted within the analytical scope of this platform. (Lines 185–188)

Revised text (excerpt): [See revised manuscript, Discussion Section 4.3, Lines 381–387].

 

Comment 5 (Major): Temper biomarker language and specify future studies

Reviewer comment: Temper biomarker language ... distinguish hypothesis-generating signatures ... specify future studies (external validation, longitudinal tracking, intervention).

Response: Thank you. In the Abstract conclusion and in the Discussion (Implications), we clarified that the present observations are hypothesis-generating and that any clinical translation would require replication in an external cohort, longitudinal tracking, and assessment of responsiveness in intervention studies. We also adjusted language that could be interpreted as definitive “biomarkers” to more cautious phrasing.

Location of changes: Abstract (Conclusions) / Discussion (Section 4.6).

Revised text (excerpt): If replicated in larger cohorts, such signatures—potentially alongside BDNF—should be regarded as hypothesis-generating; biomarker development would require external validation, longitudinal tracking, and assessment of intervention responsiveness before any clinical use is considered. (Lines 35–38)

Revised text (excerpt): [See revised manuscript, Discussion Section 4.6: Implications and future directions].

 

3. Additional Minor Comments (Terminology, Figure, and Formatting)

Comment 1. Terminology consistency

Reviewer comment: Please ensure consistent use of “3-hydroxybutyric acid” vs. “β-hydroxybutyrate” throughout the manuscript.

Response: We revised the manuscript to ensure terminology consistency and unified the term to “3-hydroxybutyric acid” throughout.

Comment 2. Figure and table clarity

Reviewer comment: I suggest to consider adding a schematic or pathway-based summary figure integrating amino acid, ketone, purine metabolism, and BDNF for reader accessibility.

Response: To improve reader accessibility, we added a pathway-based schematic integrating the amino-acid/one-carbon, ketone-body, purine/energy-related findings and BDNF as Supplementary Figure S7, and we now cite this figure in the Discussion.

Location of changes: Supplementary Figure S7 (new) / Discussion (Section 4.6).

Revised text (excerpt): For reader accessibility, a pathway-based schematic integrating the amino-acid/one-carbon, ketone-body, purine/energy-related findings and BDNF is provided in Supplementary Figure S7. (Lines 467–469)

Comment 3. Grammar and formatting

Reviewer comment: Minor typographical inconsistencies (e.g., spacing, symbols in tables) should be corrected during copyediting.

Response: We carefully reviewed the entire manuscript during revision and corrected minor typographical and formatting inconsistencies (e.g., spacing and symbols in tables) where applicable.

 

Conclusion

We believe that these revisions address the reviewers’ concerns (clarification of the statistical framework and exploratory interpretation, potential confounding, positioning of correlation analyses, platform-dependent interpretation, more cautious biomarker language, and improved reader accessibility) with minimal additional text while improving clarity. We greatly appreciate your continued guidance.

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

I thank the authors for the revisions. Previous ambiguities have been resolved; therefore, I recommend the manuscript for publication.