The Association of Salivary Conductivity with Cardiomegaly in Hemodialysis Patients

Round 1

Reviewer 1 Report

Thanks for recommending me as a reviewer. This cross-sectional study aims to explore the association between salivary conductivity and Cardiothoracic ratio(CTR). If the author completes the revision, the quality of the study will be further improved.

1. The introduction section is well written. However, readers (In particular, if the reader is a researcher in the field of applied science) with no background in CTR are likely to not understand the theoretical background . If the authors describe the theoretical background and research purpose in more detail in the introduction section, it can help the reader's understanding.

2. line 90-98: Authors should be more specific about the subject of their study. For example, the sampling of the sample, what to exclude from the study, and what to include in the study should be specifically described.

3. Figure 1 is well represented. However, the title is a bit ambiguous. I need to fix it.

4.  line 129- 145: Authors should be more specific about the scales used in this study. For example, if you add the reliability of the measures in the previous study (or at the time of development) and the reliability of the measures in this study, it can help the reader's understanding.

5. line 150-151: The authors should describe in more detail the method of testing the receiver operating characteristic (ROC) curve in this study.

6. In this study, AUC is 0.626, which is poor classification performance. The authors should add the limitations of the study in the discussion. In addition, authors should further describe methods for improving classification/prediction performance for readers.

Author Response

Rebuttal Letter

A-T Lee, Y-P Lu and C-H Chen, et al. The Association of Salivary Conductivity with Cardiomegaly in Hemodialysis Patients

Reviewer 1

Comments and Suggestions for Authors

Thanks for recommending me as a reviewer. This cross-sectional study aims to explore the association between salivary conductivity and Cardiothoracic ratio (CTR). If the author completes the revision, the quality of the study will be further improved. 

1. The introduction section is well written. However, readers (In particular, if the reader is a researcher in the field of applied science) with no background in CTR are likely to not understand the theoretical background. If the authors describe the theoretical background and research purpose in more detail in the introduction section, it can help the reader's understanding.

Response: We thank the reviewer for this comment. We have added more detailed information about CTR in the introduction section for better understanding of the theoretical background. (Page 2, line 58-67 in the section of Introduction)

2. line 90-98: Authors should be more specific about the subject of their study. For example, the sampling of the sample, what to exclude from the study, and what to include in the study should be specifically described.

Response: Thanks for your recommendation. We have specifically described the flow of patient enrollment in this study. Briefly, we recruited all hemodialysis patients in our dialysis center initially. Patients receiving dialysis less than 3 months or with active illness were excluded. Among 165 enrolled patients, 36 of whom did not complete this study due to no available Chest X ray or no saliva collection. This study was conducted in November 2018 and 129 patients were included int he final analysis. (Page 2, line 95-106 and Figure 1)

3. Figure 1 is well represented. However, the title is a bit ambiguous. I need to fix it.

 Response: We thank the reviewer for point this out. We have changed the title to “Flow chart of patient enrollment”. (Page 3, line 114)

4. line 129- 145: Authors should be more specific about the scales used in this study. For example, if you add the reliability of the measures in the previous study (or at the time of development) and the reliability of the measures in this study, it can help the reader's understanding.

Response: We thank the reviewer for this suggestion. We have provided the test-retest and internal reliability data of the thirst intensity scale tools by interclass correlation coefficient and Cronbach’s alpha coefficient. (Page 5, line 152-156)

5. line 150-151: The authors should describe in more detail the method of testing the receiver operating characteristic (ROC) curve in this study.

Response: Thanks for your recommendation. We have described the Receiver operating characteristic (ROC) curve method in more detail. (Page 5, line 172-178)

6. In this study, AUC is 0.626, which is poor classification performance. The authors should add the limitations of the study in the discussion. In addition, authors should further describe methods for improving classification/prediction performance for readers.

Response: We thank the reviewer for the comment. We have added this limitation in the section of discussion. (Page 13, line 347-350) To improve the diagnostic performance, we further combined salivary conductivity with serum osmolality, sodium, or history of diabetes mellitus in the prediction model. The combination of salivary conductivity, serum sodium and history of diabetes mellitus showed a significant increase of AUROC (0.782, 95% CI: 0.696 – 0.868) (Page 7, line 231-235). The results are also shown in the figure 4. (Page 9, line 243)

Author Response File: Author Response.docx

Reviewer 2 Report

This is a cross-sectional study aiming to explore the association between salivary conductivity and CTR. It was reported that CTR was negatively associated with salivary conductivity or serum sodium level and positively correlated with older age. The ROC for the diagnostic ability of salivary conductivity on cardiomegaly reported an AUC of 0.626 (95% CI: 0.521-0.730, P=0.02). The manuscript although interesting requires further refinement to be considered for publication:

1. In the collection and analysis of saliva, authors mention that 100 μL of saliva sample was diluted with 400 μL of deionized water. Please define the conductivity/resistivity of the deionized water that was utilized for the dilution. Ultra-pure water should be utilized for this purpose. It is mentioned that salivary conductivity was measured 3 times for each subject and the averaged values were reported. Please provide the values of all three measurements as supplementary material along with the standard deviation and the coefficient of variation.
2. In the statistical analysis section authors mention that in order to compare the difference between 2 groups, the independent t test was applied for quantitative variables. T-Test is used when there is normal distribution of the values. Were the quantitaive variables normally distributted? If not, a non parametric test such as Mann-Whitney test would be much more appropriate for the statistical analysis.
3. In the results section, in the demographics part, it is reported that patients with cardiomegaly had lower serum osmolality, lower serum sodium level and lower salivary conductivity at statistical significant level (P<0.05). However, the mean values among patients with cardiomegaly compared to those with non cardiomegaly were quite close for serum osmolality (307.83 ± 11.28 vs 311.56 ± 9.72 mOsm/kg H2O) and serum sodium level (135.88 ± 4.01 vs 137.25 ± 3.27 mEq/L). Is this difference sufficient to define subjects with cardiomegaly or it will change if utilizing a more balanced sample size among the 2 groups (cardiomegaly vs non cardiomegaly)?
4. In the results section, demographics part it is mentioned that the patients were aged 61.31 ± 11.88 years. This is not validated in Table 1 since non of the 3 values reported for age are in agreement with the above value. Please correct as applicable.
5. In Table 1, the salivary conductivity values have big standard deviations (almost 30% of the mean value). This is not usual when statistical significance is reported. In this case sd usually ranges from 10-15% of the mean value. Please provide all values for each group along with the CV as a supplementary file.
6. In Figure 3 patients were divided into low, medium and high salivary conductivity groups. Please define the range values for each group.
7. In figure 4 the AUC at the specific CI is relatively low (0.626). Have the authors tried to combine salivary conductivity with other factors (such as serum osmolality) to check if they receive an improved AUC?
8. How many of the 42 patients with cardiomegaly had lower salivary conductivity? The sample size was sufficient (N=129) but the group distribution was quite un-balanced (42 patients with cardiomegaly and 87 without cardiomegaly). Is it possible that some of the results reported might be skewed due to this unequal sample size? Please elaborate on this.
9. Since CTR is routinely used to define cardiomegaly and CVD risk on patients undergoing maintenance hemodialysis, what is the added value of the salivary conductivity? Is it a more accurate measurement? Is it easier or is it better correlated with CVD risk compared to CTR?

Author Response

Rebuttal Letter

A-T Lee, Y-P Lu and C-H Chen, et al. The Association of Salivary Conductivity with Cardiomegaly in Hemodialysis Patients

Reviewer 2

Comments and Suggestions for Authors

This is a cross-sectional study aiming to explore the association between salivary conductivity and CTR. It was reported that CTR was negatively associated with salivary conductivity or serum sodium level and positively correlated with older age. The ROC for the diagnostic ability of salivary conductivity on cardiomegaly reported an AUC of 0.626 (95% CI: 0.521-0.730, P=0.02). The manuscript although interesting requires further refinement to be considered for publication:

1. In the collection and analysis of saliva, authors mention that 100 μL of saliva sample was diluted with 400 μL of deionized water. Please define the conductivity/resistivity of the deionized water that was utilized for the dilution. Ultra-pure water should be utilized for this purpose. It is mentioned that salivary conductivity was measured 3 times for each subject and the averaged values were reported. Please provide the values of all three measurements as supplementary material along with the standard deviation and the coefficient of variation.

Response: We thank the reviewer for the comment. We used ultrapure rather than merely deionized water for the dilution of saliva. We have corrected this error and provided the conductivity/resistivity of the dilution water utilized in this study. (The conductivity is 0.055 µSiemens/cm of the water, while the resistivity is 18.2 MΩ × cm at 25°C) (Page 4, line 135-136) We have also provided the values of all three measurements of salivary conductivity in the supplemental materials (supplemental materials, Table S1)

2. In the statistical analysis section authors mention that in order to compare the difference between 2 groups, the independent t test was applied for quantitative variables. T-Test is used when there is normal distribution of the values. Were the quantitaive variables normally distributted? If not, a non parametric test such as Mann-Whitney test would be much more appropriate for the statistical analysis.

Response: We thank the reviewer for the recommendation. We have used the Kolmogorov-Smirnov method to test the normal distribution of numerical values. We also revised the statistical methods (Page 5, line 166-169) and results (Page 5, Table 1). The results of statistical significance are not changed.

3. In the results section, in the demographics part, it is reported that patients with cardiomegaly had lower serum osmolality, lower serum sodium level and lower salivary conductivity at statistical significant level (P<0.05). However, the mean values among patients with cardiomegaly compared to those with non cardiomegaly were quite close for serum osmolality (307.83 ± 11.28 vs 311.56 ± 9.72 mOsm/kg H2O) and serum sodium level (135.88 ± 4.01 vs 137.25 ± 3.27 mEq/L). Is this difference sufficient to define subjects with cardiomegaly or it will change if utilizing a more balanced sample size among the 2 groups (cardiomegaly vs non cardiomegaly)?

Response: We thank the reviewer for the comment. We have used age- and sex-matched controls (n=42 under 1:1 matching) for further sensitivity analysis. The results are quite similar as the original analysis. (Supplemental Materials, table S2)

4. In the results section, demographics part it is mentioned that the patients were aged 61.31 ± 11.88 years. This is not validated in Table 1 since non of the 3 values reported for age are in agreement with the above value. Please correct as applicable.

Response: We have corrected this error as the reviewer’s comment. (Page 5, line 188)

5. In Table 1, the salivary conductivity values have big standard deviations (almost 30% of the mean value). This is not usual when statistical significance is reported. In this case sd usually ranges from 10-15% of the mean value. Please provide all values for each group along with the CV as a supplementary file.

Response: We thank the reviewer for the comment. We have provided all values of salivary conductivity along with CV for each group in the supplemental materials (supplemental materials, Table S1)

6. In Figure 3 patients were divided into low, medium and high salivary conductivity groups. Please define the range values for each group.

Response: We have defined the range values for each group accordingly. (Page 8, line 239)

7. In figure 4 the AUC at the specific CI is relatively low (0.626). Have the authors tried to combine salivary conductivity with other factors (such as serum osmolality) to check if they receive an improved AUC?

Response: We thank the reviewer for the suggestion. To improve the diagnostic performance, we further combined salivary conductivity with serum osmolality, sodium, and history of diabetes mellitus in the prediction model. The combination of salivary conductivity, serum sodium and history of diabetes mellitus showed a significant increase of AUROC (0.782, 95% CI: 0.696 – 0.868) (Page 7, line 231-235). The results are also shown in the figure 4. (Page 9, line 243)

8. How many of the 42 patients with cardiomegaly had lower salivary conductivity? The sample size was sufficient (N=129) but the group distribution was quite un-balanced (42 patients with cardiomegaly and 87 without cardiomegaly). Is it possible that some of the results reported might be skewed due to this unequal sample size? Please elaborate on this.

Response: Thanks for your comments. We have performed 1:1 age- and sex-matched analysis to clarifying the effect of unequal sample size. The results are quite similar as the original analysis. (Supplemental Materials, table S2)

9. Since CTR is routinely used to define cardiomegaly and CVD risk on patients undergoing maintenance hemodialysis, what is the added value of the salivary conductivity? Is it a more accurate measurement? Is it easier or is it better correlated with CVD risk compared to CTR?

Response: We thank the reviewer for this important comment. Although CTR is routinely used to define cardiomegaly, it is usually measured by chest X ray, computed tomography, or echocardiography. These diagnostic tools are ionizing radiation, relatively expensive and inconvenient when compared with the salivary examination. However, the current results are still not enough to confirm that saliva conductivity is a good biomarker of cardiovascular prognosis. Therefore, we still need to conduct a prospective cohort study to further clarify the ability of saliva conductivity for predicting cardiovascular risk. We have added this viewpoint into the section of conclusion. (Page 13, line 353-363)

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

The authors faithfully completed the revision.

Author Response

Thanks very much for your kind work and consideration on publication of our paper. On behalf of my co-authors, I would like to express my great appreciation to the reviewers for their positive and constructive comments.

Thank you and best regards,

Yours sincerely.

Reviewer 2 Report

The authors have adequately adressed the comments raised.

Author Response

Thanks very much for your kind work and consideration on publication of our paper. On behalf of my co-authors, I would like to express my great appreciation to the reviewers for their positive and constructive comments.

Thank you and best regards,

Yours sincerely.