Disruption of the Functional Activity of Neutrophil Granulocytes as a Risk Factor for the Development of Lung Damage in Pregnant Women with COVID-19

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

I reviewed the cimb-2800906 manuscript. In my opinion there is a major ethic concern about the methodology of this study; The authors included in the study 54 pregnant women with moderate COVID-19 and all of them underwent a CT scan. In the majority of COVID-19 international guidelines CT scan is not recommended rutinely for patients with moderate COVID-19. Thus, CT scan in pregnant women for research purposes comprises a major ethical violation. The authors state in the methods section that "CT scans for the main group were conducted following the Methodological Recommendations for Organizing Medical Care for 123 Pregnant Women, Parturients, Maternity Patients, and Newborns with New Coronavirus Infection (COVID-19)" (version 5, 28.12.2021) [15]. However, the reference no. 15 examines the pathogenesis and management of cytokine storm and does not provide clues for the management of the pregnant women with SARS-CoV-2 infection.

Another significant methodological error is, in my opinion, that COVID-19 diagnosis was not based solely on PCR testing; rather  it was based only in clinical features in a proportion of the participants.

Comments on the Quality of English Language

Minor English language editing is required

Author Response

Response to review in attached file

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

Disruption of the Functional Activity of Neutrophil Granulocytes as a Risk Factor for the Development of Lung Damage in Pregnant Women with COVID-19

The authors present an interesting analysis of neutrophil granulocytes and COVID-19 in pregnant women. The English is of a good standard and few issues were detected. The Figures would benefit from improving, and there are some topics that could usefully be added to Methods and Discussion. More detailed comments are given below.

(1) Figure 1 is not presented well. The grey borders around the chart make it look copy-pasted, it is a multi-part chart and each part should be labelled (a), (b), (c) or similar, and the legend does not provide enough information. The axes are poorly labelled, % of what for example? % of total? % of control levels? This Figure needs substantial improvement and I suggest the authors look for 'best in class' examples from the journal and follow those standards.

(2) Figures 2-4 require a scale

(3) Figure 5, same comments as Figure 1

(4) Figure 6, no scale again

(5) Figure 7, % of what on the y-axis? I know it is % increase from line 350 but Figure axes should be self-contained

(6) I am confused by Figure 8. Are the authors proposing one of these features, or both, as their indicators? Where is the ROC curve for both features? Logistic regression can create an indicator with more than one feature (it is a multivariate tool), so surely the interesting ROC curve is for a panel with both features in it?

(7) Same comment goes to Table 4. Also, more generally, sensitivity and specificity of 69% and 60% is not 'good' in many settings, especially as there will be a confidence interval around these values. Hence the point about what would be the values for a combined test using both features?

(8) The Methods section is clear, but I cannot find any statement on data availability. What data do the authors intend to make available for the purposes of reproducibility and have they followed both journal guidelines and research best practices in this?

(9) As part of the Discussion, it would be helpful to mention treatment options related to this issue. Specifically, the authors discuss immature neutrophils and their role (lines 71, 397) but do not mention dexamethasone and its relationship with neutrophils, highly relevant and discussed in the literature (PMID: 34782790 and PMID: 36292938). This should be set in the wider context of corticosteroids in pregnancy (PMID: 26590298 and PMID: 32769659). What other treatment options are available - in other words, what would be the clinical response to the tests that the authors propose? How would these findings influence current recommendations around corticosteroids?

(10) It might be helpful in Table 4 to be explicit that the confidence interval and p-value applies to the ROC value. Does the p-value capture the likelihood that the ROC value is not .50? This is not particularly clear at present.

(11) I cannot find a limitations paragraph in Discussion. It is always helpful for the authors to highlight the limitations of their work, in order to help the reader put it into context. For example, the authors state that "a multivariate regression analysis" was conducted, but presumably no test:train splitting was done. This can lead to overstatement overfitting of ROC values (see PMID: 34862523 and PMID: 35316964). Whilst sometimes unavoidable, it is the kind of issue that should be flagged under limitations, alongside any limitations related to statistical power, data on medication, and so on.

(12) Finally, somewhat out of sequence, I am confused by what cohorts have actually been analysed. For example, the paper at one point states that "The main group included 54 women in the third trimester of pregnancy with moderate COVID-19 severity" and that "Subgroup 1 (n=43) included pregnant women with minimal lung changes (less than 25%, CT-1), and subgroup 2 included those with moderate lung changes (25-50%, CT-2)"

Then, "The control group comprised 35 uninfected SARS-CoV-2 women with uncomplicated pregnancies [with] absence of respiratory infections during the current pregnancy"

How can I reconcile these cohorts to Table 1 - what cohort do the 11 women with "SARS-CoV-2 Uninfected" status belong to?

Comparing this to Table 2 is also confusing, as we have the same 43 / 11 subgroup split, but the labelling is different and inconsistent with the text. Table 2 contrasts "minimal" and "average", but the text refers to "minimal" and "moderate" in line 105. Is the average value "moderate"? How has "average" been defined in this context? Is this the same 11 as Table 1, in which case Table 1 is probably wrong? These issues should be made much clearer to the reader to avoid confusion.

Author Response

Response to review in attached file

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

The topic of paper is of great interest. The paper is broad and detailed.

In general, the paper is well structured and facilitates the reading and the follow-up of the study. Methodologically the study is well supported and explained. However, some minor clarifications must be included to accept their publication.

The abstract needs to be improved. The objectives of the study, a further explanation of the methodology and a discussion section should be included.

The introduction section is well structured. However, it should end with a description of the objectives of the study.

Materials and methods. It is very well explained.  

Results. It is well explained, however, some of the comments should be removed from this section and included in the discussion section. For example, lines 271 to 274, lines 311 to 314, and lines 324 to 330.

Discussion. It is correct, well written and must include the explanations of the results obtained, mentioned above.

The last paragraph should be included in the discussion.

Conclusions. This section should be expanded to include a more robust explanation of the results obtained.

Author Response

Response to review in attached file

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

The authors have responded to the comments made. There are still some minor issues to be addressed in final editing, for example line 488 "[25, 26].Destabilization" is missing a space, and there are some other minor errors. The charts are improved versus the previous version, but Figures 5 and 7 appear to have some wrong labelling on the y-axis which should be amended.

I think it is disappointing that no effort to make the underlying data available has been made. This is important for reproducibility. Given journal requirements and best practice, the authors should either reconsider this and upload to a repository, OR include a data availability statement that explicitly says that the data will not be made available.

Journal requirements are as follows, for the authors' convenience:

"MDPI Research Data Policies

MDPI is committed to supporting open scientific exchange and enabling our authors to achieve best practices in sharing and archiving research data. We encourage all authors of articles published in MDPI journals to share their research data. Individual journal guidelines can be found at the journal ‘Instructions for Authors’ page. Data sharing policies concern the minimal dataset that supports the central findings of a published study. Generated data should be publicly available and cited in accordance with journal guidelines.

MDPI data policies are informed by TOP Guidelines and FAIR Principles."

Comments on the Quality of English Language

No additional comments.

Author Response

Reply to review in attachment            

Author Response File: Author Response.pdf