Association of Inflammatory–Hematological Biomarkers with Hypertension and Related Comorbidities

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

This manuscript addresses an important and clinically relevant topic, namely the role of inflammatory-hematological biomarkers in patients with hypertension and frequent metabolic and renal comorbidities. The large dataset and the effort invested in statistical analysis are evident. However, several methodological and interpretative aspects require clarification and refinement to ensure that the conclusions are fully supported by the presented data.

Main comments

1. Interpretation of predictive value and ROC analyses
   While multiple ROC analyses are statistically significant, most biomarkers demonstrate low specificity, often below clinically acceptable thresholds. This substantially limits their practical predictive utility. The conclusions currently overemphasize “predictive” capacity and should be revised to reflect an associative or exploratory role rather than a robust diagnostic or prognostic one.

2. Definition of “prediction” versus association
   Throughout the manuscript, the terms “predictive” and “diagnostic” are used interchangeably with statistically significant associations. Given the retrospective design and limited discrimination (AUC values frequently <0.6), the authors should clearly distinguish between statistical association and true predictive performance.

3. Patient grouping and disease characterization
   The grouping strategy (HTN-only, HTN+CKD, HTN+T2DM, HTN+CKD+T2DM) is reasonable. However, the lack of disease staging (e.g., CKD stages, HTN severity, glycemic control) limits interpretability. This limitation should be more clearly acknowledged, and conclusions adjusted accordingly.

4. Clinical relevance of identified cut-off values
   Several proposed cut-off values yield high sensitivity but near-zero specificity. These thresholds are unlikely to be clinically actionable. The authors should discuss this limitation more explicitly and avoid implying immediate applicability in clinical risk stratification.

5. Overextension of conclusions
   Statements suggesting that these biomarkers could be used for early diagnosis, monitoring, or prevention should be tempered. The data support hypothesis generation rather than implementation into clinical algorithms at this stage.

Additional comments

1. Length and structure of the Introduction
   The Introduction is comprehensive but somewhat lengthy and repetitive. Streamlining the background and focusing on the specific knowledge gap would improve readability.

2. Results presentation
   The Results section is extensive and statistically dense. Condensing some repetitive descriptions and emphasizing clinically meaningful findings would improve clarity.

3. Figures and tables
   Several ROC figures appear redundant when compared with tabulated data. Consider reducing the number of figures or moving some to the Supplementary Materials.

4. Language and terminology
   The manuscript contains frequent grammatical inaccuracies and awkward phrasing. Professional language editing is recommended to improve clarity and consistency. Terminology (e.g., “predictive,” “diagnostic,” “prognostic”) should be used more precisely throughout.

5. Limitations section
   While limitations are acknowledged, they could be expanded slightly to include the impact of the retrospective design, the lack of ambulatory blood pressure data, and the absence of adjustment for inflammatory confounders beyond exclusion criteria.

Overall, this study provides valuable observational data from a large cohort and contributes to the growing literature on inflammation in hypertension and cardiometabolic disease. With a more cautious interpretation of ROC findings, a clearer distinction between association and prediction, and refinement of conclusions, the manuscript could be significantly strengthened and offer meaningful insights for clinicians and researchers.

Comments on the Quality of English Language

The manuscript is generally understandable. However, it contains frequent grammatical inaccuracies, awkward sentence structures, and inconsistent terminology. Careful language editing by a proficient English speaker or a professional editing service is recommended to improve clarity, readability, and precision, particularly in the Methods, Results, and Discussion sections.

Author Response

Dear Reviewer,  

 

We appreciate your time in evaluating our work and the thorough review you provided. We agree with these observations; accordingly, we have considered each point and made changes accordingly.

Main comments

1. Interpretation of predictive value and ROC analyses
   While multiple ROC analyses are statistically significant, most biomarkers demonstrate low specificity, often below clinically acceptable thresholds. This substantially limits their practical predictive utility. The conclusions currently overemphasize “predictive” capacity and should be revised to reflect an associative or exploratory role rather than a robust diagnostic or prognostic one.

Thank you for this observation. We agree and therefore have introduced in the Materials and Methods section the description of ROC Statistical Significance, and AUC. We corrected the interpretation in the manuscript, step by step, and excluded the idea of “predictive”given the statistical results.

2. Definition of “prediction” versus association
   Throughout the manuscript, the terms “predictive” and “diagnostic” are used interchangeably with statistically significant associations. Given the retrospective design and limited discrimination (AUC values frequently <0.6), the authors should clearly distinguish between statistical association and true predictive performance.

We agree with this recommendation; to avoid misleading and to correct the interpretation we have made, we have made changes to our results, discussion, and conclusions.

3. Patient grouping and disease characterization
   The grouping strategy (HTN-only, HTN+CKD, HTN+T2DM, HTN+CKD+T2DM) is reasonable. However, the lack of disease staging (e.g., CKD stages, HTN severity, glycemic control) limits interpretability. This limitation should be more clearly acknowledged, and conclusions adjusted accordingly.

We have rephrased the study limitation, included it, and revised the conclusions accordingly; please verify.

4. Clinical relevance of identified cut-off values
   Several proposed cut-off values yield high sensitivity but near-zero specificity. These thresholds are unlikely to be clinically actionable. The authors should discuss this limitation more explicitly and avoid implying immediate applicability in clinical risk stratification.

We have included this aspect at the end of 4. Discussion, indeed, this needed to be highlighted for future readers as it is a crucial and correct key finding in this research.

5. Overextension of conclusions
   Statements suggesting that these biomarkers could be used for early diagnosis, monitoring, or prevention should be tempered. The data support hypothesis generation rather than implementation into clinical algorithms at this stage.

We have rephrased our conclusion; our results reflect an association that further generates a hypothesis, rather than implementation into clinical algorithms at this stage. Thank you for this recommendation.

Additional comments

1. Length and structure of the Introduction
   The Introduction is comprehensive but somewhat lengthy and repetitive. Streamlining the background and focusing on the specific knowledge gap would improve readability.

The introduction has been revised by adding additional information and rephrasing paragraphs to improve the clarity of the manuscript.

2. Results presentation
   The Results section is extensive and statistically dense. Condensing some repetitive descriptions and emphasizing clinically meaningful findings would improve clarity.

We have reduced the statistical data presented in the main text as much as possible, since they are also shown in figures and tables, and we have tried to comment on each result while emphasizing its clinical significance.

3. Figures and tables
   Several ROC figures appear redundant when compared with tabulated data. Consider reducing the number of figures or moving some to the Supplementary Materials.

We have reduced the number of figures in the main manuscript, for example, Figure 2. And 3, and moved them to the Supplementary Materials. We have kept the other 2 figures regarding ROC analysis, in the main manuscript, but if necessary we can move them

4. Language and terminology
   The manuscript contains frequent grammatical inaccuracies and awkward phrasing. Professional language editing is recommended to improve clarity and consistency. Terminology (e.g., “predictive,” “diagnostic,” “prognostic”) should be used more precisely throughout.

We have made as many grammatical corrections and rephrased paragraphs and sections as possible to improve clarity and consistency in the manuscript.

5. Limitations section
   While limitations are acknowledged, they could be expanded slightly to include the impact of the retrospective design, the lack of ambulatory blood pressure data, and the absence of adjustment for inflammatory confounders beyond exclusion criteria.

We have introduced limitations on the elements mentioned. Hope we have improved this section, please verify.

 

We have revised the entire manuscript, corrected grammar, rephrased where needed, and hopefully improved clarity. To further raise the quality of our paper, we will use the Journal's professional Language help in the next steps. Your valuable comments helped to improve our original research.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

The manuscript titled” Value of inflammatory-hematological biomarkers in predicting HTN in patients with comorbidities  “  addresses the role of blood cells (NLR, PLR, neutrophils, lymphocytes, platelets) across patients with hypertension and those with type 2 diabetes mellitus and/or chronic kidney disease while the secondary objective was to explore clinically relevant cut-off values for these markers. The topic is important and offers valuable guidance for relevant to cardiovascular risk.

However, the following revisions are required .

• In diagnostic value studies, AUC values above 0.90 are interpreted as indicating a very good diagnostic performance of the test, while AUC values below 0.80, even if they are statistically significant, are interpreted as indicating a very limited clinical usability of the test. I recommended to use association or differentiation rather than predication because AUC values in the current study are below 0.6, which indicates poor discrimination, not predictive ability
• The authors are encouraged to revise the manuscript title. A suggested alternative “association of inflammatory–hematological biomarkers with hypertension and related comorbidities”
• In the Abstract, the formatting of p-values and AUC should be standardized. For example,” p- 000*”* should be corrected to p <0.001** , “AUC == 0.442” should be corrected to “AUC = 0.442” (line 29), and “AUC 0.567” should be revised to “AUC = 0.567” (line 30)
• The Introduction would benefit from explanation of how inflammation and immune dysregulation within the mosaic theory of hypertension validate the investigation of circulating leukocyte-based biomarkers and platelet-based biomarkers.
• Statistical Analysis: what is the optimal cut-off and AUC values with 95% confidence intervals should be reported to support diagnostic
• The data are valuable; however, terminology must be corrected, and claims must be consistently aligned with ROC performance standards. In the Results section, AUC values below 0.6 should not be interpreted as predictive or prognostic. The authors are encouraged to replace such terminology with discriminatory or association or differentiation or similar wording.
• Figures 2 and 3 are presented in parallel; however, their legends are combined and placed below the figures. For clarity with standard journal formatting, each figure should be presented separately and accompanied by its own legend. The same is true for Figure 4 and 5. The authors are requested to revise the figure layout accordingly
•  

Comments on the Quality of English Language

• Professional language editing is recommended.

Author Response

Dear Reviewer,

 

We want to thank you, on behalf of our team, for your time in reviewing our original research and for your helpful observations and comments. We have considered each recommendation and made the necessary changes. Please verify each point.

• In diagnostic value studies, AUC values above 0.90 are interpreted as indicating a very good diagnostic performance of the test, while AUC values below 0.80, even if they are statistically significant, are interpreted as indicating a very limited clinical usability of the test. I recommended to use association or differentiation rather than predication because AUC values in the current study are below 0.6, which indicates poor discrimination, not predictive ability

 

Thank you for this observation. We agree with this statement regarding the statistical analysis and its interpretation. We have described in Chapter 2.3. Statistical Analysis: the interpretation of AUC values and made corrections in the manuscript, according to the AUC values.

 

• The authors are encouraged to revise the manuscript title. A suggested alternative “association of inflammatory–hematological biomarkers with hypertension and related comorbidities”

 

Thank you for this suggestion. We agree and have made changes accordingly.

 

• In the Abstract, the formatting of p-values and AUC should be standardized. For example,” p- 000*”* should be corrected to p <0.001** , “AUC == 0.442” should be corrected to “AUC = 0.442” (line 29), and “AUC 0.567” should be revised to “AUC = 0.567” (line 30).

 

We have corrected the standardization formatting of p-values and AUC in the abstract and in the manuscript.

 

• The Introduction would benefit from explanation of how inflammation and immune dysregulation within the mosaic theory of hypertension validate the investigation of circulating leukocyte-based biomarkers and platelet-based biomarkers.

 

We have rephrased the introduction as recommended and included a paragraph describing the connection between inflammation and immune dysregulation within the mosaic theory of hypertension and the investigation of circulating leukocyte- and platelet-based biomarkers.

 

• Statistical Analysis: what is the optimal cut-off and AUC values with 95% confidence intervals should be reported to support diagnostic.

 

We have introduced the optimal cutoff values in Materials and Methods and described the aspects of ROC Statistical Significance and AUC.

 

• The data are valuable; however, terminology must be corrected, and claims must be consistently aligned with ROC performance standards. In the Results section, AUC values below 0.6 should not be interpreted as predictive or prognostic. The authors are encouraged to replace such terminology with discriminatory or association or differentiation or similar wording.

 

We agree with this observation as well, we have corrected with the correct terminology corresponding to AUC values and by ROC performance standards. 

 

• Figures 2 and 3 are presented in parallel; however, their legends are combined and placed below the figures. For clarity with standard journal formatting, each figure should be presented separately and accompanied by its own legend. The same is true for Figure 4 and 5. The authors are requested to revise the figure layout accordingly

 

We have moved Figures 2 and 3 into Supplementary as advised, and changed the layout of the remaining Figures 4 and 5;

 

Thank you once again. We believe that your valuable comments helped to improve the quality of our original research.

 

 

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

The authors have made a visible effort to revise the manuscript and address several of the comments raised during the previous review round. The structure has improved, and additional analyses and clarifications have been incorporated. Nevertheless, important methodological, statistical, and editorial issues remain and should be further addressed.

The Introduction remains overly long and repetitive, with multiple restatements of similar concepts related to inflammation, endothelial dysfunction, and hematological markers. The authors are encouraged to streamline this section and to focus more clearly on the specific clinical and scientific gap that the present study aims to address.

Although the retrospective, single-center design is acknowledged, several potential sources of bias remain insufficiently discussed. In particular, information regarding antihypertensive, antidiabetic, lipid-lowering, and anti-inflammatory therapies is lacking, despite their well-known influence on inflammatory indices. A more detailed description of medication use and comorbidity management would substantially strengthen the validity of the findings.

The statistical methodology still requires further clarification. The rationale for variable selection in regression models is not adequately explained, and important confounders such as age, sex, renal function, anemia, obesity, and dyslipidemia are not consistently controlled. In addition, the issue of multiple testing should be more explicitly addressed.

The presentation of results remains overly extensive, with substantial redundancy between text, tables, and figures. Several sections mainly restate numerical data without sufficient clinical interpretation. Moreover, the very small size of some subgroups, particularly the HTN+CKD+T2DM group, substantially limits the reliability of subgroup analyses and should be more cautiously discussed.

The ROC analyses are carefully performed; however, most identified cut-off values are characterized by low specificity and limited discriminatory performance. Therefore, the clinical applicability of these markers should be interpreted more conservatively. Conclusions regarding predictive or prognostic utility should be further tempered and aligned more closely with the actual statistical performance.

The language and style of the manuscript have improved compared with the previous version, but substantial grammatical, syntactic, and stylistic problems persist. Professional language editing is strongly recommended to improve clarity and readability.

In summary, this study addresses a clinically relevant topic and is based on a large patient cohort. However, further refinement of methodology, statistical analysis, interpretation, and language is necessary to ensure scientific rigor and clinical relevance.

Comments on the Quality of English Language

The manuscript has improved linguistically compared with the previous version; however, numerous grammatical, syntactic, and stylistic inaccuracies persist throughout the text. Several sentences remain unclear or repetitive, which affects readability and scientific clarity. Professional English language editing is strongly recommended prior to further consideration.

Author Response

Dear Reviewer,  

 

We deeply appreciate your valuable comments and have addressed them with proper attention and rigor. Therefore, we have restructured, corrected errors, and rephrased as needed. Additionally, the version of the manuscript submitted now reflects the changes made, and we have appealed to the Journal’s Professional Service for English Editing, as recommended.

 

The Introduction remains overly long and repetitive, with multiple restatements of similar concepts related to inflammation, endothelial dysfunction, and hematological markers. The authors are encouraged to streamline this section and to focus more clearly on the specific clinical and scientific gap that the present study aims to address.

We initially introduced additional inflammatory aspects to address the other reviewers' comments; however, for clarity, we simplified the inflammatory panel data and retained only the key elements, as the pro-inflammatory pathway links the pathologies investigated. We then introduced the current knowledge on NLR and PLR normal values as the “baseline” (for comparison with our results) and their associations with hypertension, diabetes, and renal disease. Most studies focus on individual diseases, and fewer examine disease associations. After extensive research, we could not identify studies that compared the 2 biomarkers across all 3 diseases. Using multivariate analysis, we could compare parameter values, assess whether there is a positive association, and establish cut-off values (based on the current data). Additionally, we found it appropriate to investigate the variations in individual neutrophils, lymphocytes, and platelets, as they underlie these ratios, to better understand the pathophysiological changes that occur in these cases.

 

Although the retrospective, single-center design is acknowledged, several potential sources of bias remain insufficiently discussed. In particular, information regarding antihypertensive, antidiabetic, lipid-lowering, and anti-inflammatory therapies is lacking, despite their well-known influence on inflammatory indices. A more detailed description of medication use and comorbidity management would substantially strengthen the validity of the findings.

 

Because this is a retrospective study based on data extracted from our hospital’s medical database, which has limitations in obtaining drug and medication information at the time of patient admission, we found it appropriate to include this important aspect, as advised, in the limitations section. We agree that patients' medications (cardiovascular and antihypertensive) influence inflammatory patterns and blood cell variations; therefore, please review the extended paragraph on this subject in the limitations section.  

 

The statistical methodology still requires further clarification. The rationale for variable selection in regression models is not adequately explained, and important confounders such as age, sex, renal function, anemia, obesity, and dyslipidemia are not consistently controlled. In addition, the issue of multiple testing should be more explicitly addressed.

We thank the reviewer for this valuable comment and the opportunity to better clarify the analytical framework.

• Study aim and analytic approach (no multivariable regression modelling). We deleted “the use of the regression model” inserted in 2.3. Statistical Analysis”-our mistake, as it was not suitable as a method for this study; not used.  

The primary objective of this study was to provide a population-level, descriptive-comparative overview of related clinical conditions (HTN, HTN+CKD, HTN+T2DM, HTN+T2DM+CKD), focusing on distributional differences in inflammatory indices and on comparisons of their ROC-derived cut-off values across groups. Accordingly, the statistical approach was designed for group comparisons (non-parametric tests) and diagnostic performance assessment (ROC analysis), rather than for building multivariable predictive regression models. Multivariable predictive regression models are more appropriate for smaller sample groups in which multiple other variables are investigated simultaneously. We have revised the manuscript to make this rationale explicit in the Statistical Analysis and Discussion sections.

• Confounding and covariate control-

We agree that age, sex, renal function, anemia, obesity, and dyslipidemia are clinically important factors that may influence both biomarker levels and disease status. However, our purpose was to identify specific cut-off values across clinically defined groups - and not necessarily to estimate specific independent effects or causal associations that would require multivariable adjustment. We have rephrased the manuscript to better clarify this aspect.

• Multiple testing: We used a post hoc test after the Kruskal-Wallis test for pairwise comparisons, with significance adjusted by the Bonferroni correction for multiple comparisons (see supplementary tables and figure). We have introduced this element in 2.3. Statistical Analysis -was missing.

 

The presentation of results remains overly extensive, with substantial redundancy between text, tables, and figures. Several sections mainly restate numerical data without sufficient clinical interpretation. Moreover, the very small size of some subgroups, particularly the HTN+CKD+T2DM group, substantially limits the reliability of subgroup analyses and should be more cautiously discussed.

 

We have revised the entire manuscript, shortened and re-adjusted each section, deleted redundant information from the main text, which is also visible in the figures and tables, and completed as much as possible with additional comments after each result (results section) and in the Discussion. We have deleted the 2 figures and retained only the figure with the ROC analysis that showed the most significant statistical results, as we believe it is important for our paper overview and for readers. We noted, where appropriate, the small group sample size and cautioned against interpreting these results. The Discussion chapter was therefore refocused; we found it more suitable to include, at first, data on inflammation and the rationale for investigating these biomarkers and the aforementioned diseases. We then highlighted the results, adjusted them in accordance with current literature, and finally discussed our limitations.

 

The ROC analyses are carefully performed; however, most identified cut-off values are characterized by low specificity and limited discriminatory performance. Therefore, the clinical applicability of these markers should be interpreted more conservatively. Conclusions regarding predictive or prognostic utility should be further tempered and aligned more closely with the actual statistical performance.

 

We have removed any fragments that might suggest our results have predictive or prognostic performance, as this is inaccurate. The conclusions have been revised to highlight our main results and the need for larger studies to validate these findings.

Author Response File: Author Response.pdf