Relationship between the Superior Attachment of the Uncinate Process and Pneumatization of the Middle Turbinate—A Radiological Study

Round 1

Reviewer 1 Report

The authors carry out a very interesting study of the anatomy of the uncinate process, the lamella of the middle turbinate and the frontal recess with clear implications in the surgical treatment of sinonasal and frontal sinus pathology. The methodology is clearly exposed with very illustrative images that greatly facilitate the reading and understanding of the study. The results presented clearly and in the discussion the results obtained are well summarized and compared with those results obtained by other authors.The bibliography is very complete. In summary, I think it is quite an appropriate article that deserves its publication.

Author Response

The authors carry out a very interesting study of the anatomy of the uncinate process, the lamella of the middle turbinate and the frontal recess with clear implications in the surgical treatment of sinonasal and frontal sinus pathology. The methodology is clearly exposed with very illustrative images that greatly facilitate the reading and understanding of the study. The results presented clearly and in the discussion the results obtained are well summarized and compared with those results obtained by other authors.The bibliography is very complete. In summary, I think it is quite an appropriate article that deserves its publication.

We appreciate the reviewer’s comment. Also, we would like to thank the time spent reading our manuscript.

Reviewer 2 Report

This manuscript draft tries to assess the relationship between the superior attachment of the uncinate process (SAUP) and the pneumatization of the middle turbinate. By analyzing 200 CT scans (379 sides), the authors subclassified the SAUP into 6 types and claimed a statistically significant association between the lateral attachment of the uncinate process and the presence of a lamellar cell. However, there remain severe problems with the manuscript, among others lack of novelty, statistical issues, and language flaws/typos. Hence, this article is not a substantial addition to the literature in this field.

 

Major flaws:

- The data regarding the prevalence of SAUP have been shown by other authors (in mostly larger cohorts). No multiple comparisons correction seems to have been applied. As the only supposedly statistically significant result is a p of 0.029, with an appropriate correction, no result will be significant any more.

- The authors do not explain what a concha bullosa is and also omit the definition of a lamellar cell, instead just their own papers with questionable classification systems are being cited. While every otolaryngologist knows what the authors are talking about, this is not a subspecialty journal and thus a proper definition for medical personnel working in other disciplines should be provided.

 

Minor remarks:

- An incredible amount of language errors and typos occur throughout the manuscript, it is advised that a native speaker corrects the draft prior to a potential resubmission. E.g., neumatization (page 1, line 23), nose1 (page 1, line 33), preoperatory (page 2, line 5), rhinosinusitis5 (page 2, line 15) etc.

- Most authors define the uncinate process as a part of the ostiomeatal complex (the authors say that it is medial of said complex – see page 1, lines 33-34).

- Figure 1 should also mention in its legend that it was modified from Landsberg and Friedman. Label the left side as panel A, the right side as panel B, describe everything in the legend. The CT scan in II is sagittal, which is different from every other image (coronal).

- Inclusion and exclusion criteria are repetitive. If the last 200 CT scans at the institute are analyzed, of course no patients without a CT scan will be included.

- Page 3, line 42: “---blinded---“. Apparently, the authors had submitted the manuscript to another journal before and forgot to edit this part as “surgeries” has an open review process.

Author Response

This manuscript draft tries to assess the relationship between the superior attachment of the uncinate process (SAUP) and the pneumatization of the middle turbinate. By analyzing 200 CT scans (379 sides), the authors subclassified the SAUP into 6 types and claimed a statistically significant association between the lateral attachment of the uncinate process and the presence of a lamellar cell. However, there remain severe problems with the manuscript, among others lack of novelty, statistical issues, and language flaws/typos. Hence, this article is not a substantial addition to the literature in this field.

Thank you for the time and effort spent reviewing our manuscript. Most of the authors are also reviewers and we understand the effort it requires.

Major flaws:

- The data regarding the prevalence of SAUP have been shown by other authors (in mostly larger cohorts). No multiple comparisons correction seems to have been applied. As the only supposedly statistically significant result is a p of 0.029, with an appropriate correction, no result will be significant any more.

Regarding the first part of the commentary, in table 2 it is summarized the results of other studies assessing the superior attachment of the uncinate process. Only one study has a larger cohort.

In relation to the second part of the commentary. Corrections for multiple comparisons are used when several groups are studied at the same time, with the ANOVA test. When we studied anterior/posterior concha bullosa and lateral/medial superior attachment of the uncinate process we only performed one comparison with the chi2 test.

- The authors do not explain what a concha bullosa is and also omit the definition of a lamellar cell, instead just their own papers with questionable classification systems are being cited. While every otolaryngologist knows what the authors are talking about, this is not a subspecialty journal and thus a proper definition for medical personnel working in other disciplines should be provided.

We agree with the reviewer, It has been explained (page 2, line 21-27).

Yes, we used our own classification of the concha bullosa, since we were studying it. It has been published in peer-reviewed journals and does not involve any treatment or diagnosis process, only a way of communicating anatomical variants. Sorry, but I do not understand why it questionable.

 Minor remarks:

- An incredible amount of language errors and typos occur throughout the manuscript, it is advised that a native speaker corrects the draft prior to a potential resubmission. E.g., neumatization (page 1, line 23), nose1 (page 1, line 33), preoperatory (page 2, line 5), rhinosinusitis5 (page 2, line 15) etc.

Thank you. We are not native speakers. The manuscript was sent originally to a professional English translator. However, after the review process, this manuscript can be sent again for a second look.

- Most authors define the uncinate process as a part of the ostiomeatal complex (the authors say that it is medial of said complex – see page 1, lines 33-34).

It has been corrected (page 1, line 33-34)

- Figure 1 should also mention in its legend that it was modified from Landsberg and Friedman. Label the left side as panel A, the right side as panel B, describe everything in the legend. The CT scan in II is sagittal, which is different from every other image (coronal).

The drawing and CT scan images are made by us. However, we have mentioned in the figure 1 legend that it was adapted from Landsberg and Friedman.

It has been labeled as A and B (left and right).

Yes, it only includes a sagittal image in the type II SAUP.It is because the attachment of the uncinate process in the agger nasi cell is better appreciated in the sagittal image rather than the coronal. The uncinate process is attached to the posterior wall of the agger nasi cell. Therefore, the coronal image showing the superior part of the uncinate process, usually will not show the whole agger nasi cell, only the posterior part. Therefore, this we believe that sagittal image is more explanatory.

- Inclusion and exclusion criteria are repetitive. If the last 200 CT scans at the institute are analyzed, of course no patients without a CT scan will be included.

It was eliminated (page 3, line 22)

- Page 3, line 42: “---blinded---“. Apparently, the authors had submitted the manuscript to another journal before and forgot to edit this part as “surgeries” has an open review process.

It has been modified (page 4, line 17).

Reviewer 3 Report

Dear Authors,

I  really appreciate your work and I consider it as an important step for the future researches in the management of chronic rinosinusitis. The chronic frontal sinusitis is a serious and a frequent pathology, and more studies with this topic could improve future surgical management.  

The paper requires same minor corrections:

• Fig. 1 needs more informations on the images: the orbit - Or, Maxillary sinus - M, Nasal septum - NS, etc.
• in the discussion section, page 5,  the paragraph noted between lines 19-22, could be reformatted like a simple affirmation and the possibles accusations could be eliminated.

-  typing errors : abstract (line 23),  introduction  (page 2, line 15), etc... 

 

Author Response

Dear Authors,

I  really appreciate your work and I consider it as an important step for the future researches in the management of chronic rinosinusitis. The chronic frontal sinusitis is a serious and a frequent pathology, and more studies with this topic could improve future surgical management.  

Thank you for your time and effort reviewing our manuscript

The paper requires same minor corrections:

• Fig. 1 needs more informations on the images: the orbit - Or, Maxillary sinus - M, Nasal septum - NS, etc.
• It has been modified.
• in the discussion section, page 5,  the paragraph noted between lines 19-22, could be reformatted like a simple affirmation and the possibles accusations could be eliminated.

We agree, it has been modified.

-  typing errors : abstract (line 23),  introduction  (page 2, line 15), etc... 

Another reviewer has appreciated some typos and language errors. We are not native speakers, and despite it has been originally reviewed by a professional English translator, it seems that some errors remained. At the end of the reviewing process, we will sent again the manuscript for a second look.

Reviewer 4 Report

The structure of the paranasal sinuses is complicated. On the other hand, in sinus surgery, it is necessary to perform the surgery safely without any secondary damage.

CT is the most important test in sinus surgery. Detailed evaluation of CT images is required before surgery. Uncinate process is a part that is always treated in sinus surgery. This paper evaluates the relationship between attachment of the uncinate process and the middle turbinate, and the attempt is interesting. 

Comments

I think the explanation of figure1 is too concise. I think it's better to show what the arrows are pointing to.

Why does only figure 1 II show a sagittal section?

Figure 2 is difficult to understand. Why not show a typical CT image of each type?

The definition of "lamellar cell" is unclear. I think it's better to state the definition clearly and show an image of a typical lamellar cell.

Minor comment

page 2 line15 : "rhinosinusitis5"  I think there is a misspelling. 
"rhinosinusitis [5]"is correct?

Author Response

The structure of the paranasal sinuses is complicated. On the other hand, in sinus surgery, it is necessary to perform the surgery safely without any secondary damage.

CT is the most important test in sinus surgery. Detailed evaluation of CT images is required before surgery. Uncinate process is a part that is always treated in sinus surgery. This paper evaluates the relationship between attachment of the uncinate process and the middle turbinate, and the attempt is interesting. 

Thank you for the time and effort spent reviewing our manuscript. Most of the authors are also reviewers and we understand the effort it requires.

Comments

I think the explanation of figure1 is too concise. I think it's better to show what the arrows are pointing to.

Thank you, it has been better explained

Why does only figure 1 II show a sagittal section?

It is because the attachment of the uncinate process in the agger nasi cell is better appreciated in the sagittal image rather than the coronal. The uncinate process is attached to the posterior wall of the agger nasi cell. Therefore, the coronal image showing the superior part of the uncinate process, usually will not show the whole agger nasi cell, only the posterior part. Therefore, this we believe that sagittal image is more explanatory.

Figure 2 is difficult to understand. Why not show a typical CT image of each type?

The definition of "lamellar cell" is unclear. I think it's better to state the definition clearly and show an image of a typical lamellar cell.

It has been explained (page 2, line 25-27)

Minor comment

page 2 line15 : "rhinosinusitis5"  I think there is a misspelling. 
"rhinosinusitis [5]"is correct?

Yes, correct

 

 

Round 2

Reviewer 2 Report

As stated in the last review, there remain severe problems with the manuscript, among others lack of novelty, statistical issues, and language flaws/typos. Hence, this article is not a substantial addition to the literature in this field.

After reading the following statement, I decided that it would not be worth the time and effort to continue with the review: "In relation to the second part of the commentary. Corrections for multiple comparisons are used when several groups are studied at the same time, with the ANOVA test. When we studied anterior/posterior concha bullosa and lateral/medial superior attachment of the uncinate process we only performed one comparison with the chi2 test." Trying to educate me about statistics when the authors clearly have no idea what they are talking about is simply outrageous. Please carefully read up on multiple comparison correction or ideally, take an extensive statistics class. This quote is from wikipedia to keep it simple: "For example, if one test is performed at the 5% level and the corresponding null hypothesis is true, there is only a 5% chance of incorrectly rejecting the null hypothesis. However, if 100 tests are each conducted at the 5% level and all corresponding null hypotheses are true, the expected number of incorrect rejections (also known as false positives or Type I errors) is 5. If the tests are statistically independent from each other, the probability of at least one incorrect rejection is approximately 99.4%." So just to make this an educational review: No, multiple comparison correction is not only applied to ANOVA tests. If you run one ANOVA and it tells you that the groups are different, that is your answer and you do not need a multiple comparison correction. Only when you then run multiple t-tests or similar comparisons to find out which groups are different from each other, you will have to correct. The same is true here. The more tests you run (yes, also chi squared tests), the likelier it is to get a statistically significant result. With an alpha level of 0.05 and at least 8 comparisons whose p values are presented in table 1, you can type these variables in on a website like this (https://www.statology.org/bonferroni-correction-calculator/) and find out that - when using Bonferroni - you need a p that is smaller than 0.00625 to be statistically significant - which your results do not have! The number even comes with an interpretation, which I cite here: "If you conduct 8 comparisons, only reject the null hypothesis of each comparison if it has a p-value less than 0.00625." I hope you have now understood the problem, this is as basic as it gets when it comes to statistics. 

Yes, it is the second largest group ever according to table 2, but what does it help if it doesn't offer new insights, adds 8 samples compared to the 2006 Ercan study, a paper published 16 (!) years ago, and doesn't even have half as many as Mahmutoğlu in 2015? But the most problematic thing is that the statistics are simply wrong (see above), which is why the paper cannot be fixed and you need a way larger sample to draw your conclusions.

Author Response

As stated in the last review, there remain severe problems with the manuscript, among others lack of novelty, statistical issues, and language flaws/typos. Hence, this article is not a substantial addition to the literature in this field.

We thank the reviewer their time spent reading our manuscript. We feel sorry that our comments have been badly received by the reviewer. Despite we do not know you, it can be seen by your comments that you are a great connoisseur of rhinology and well trained in statistics. Therefore, your comments are valid and well received by us. Please, accept our apologies and receive our comments as a valid discussion between colleagues who are trying their best.

After reading the following statement, I decided that it would not be worth the time and effort to continue with the review: "In relation to the second part of the commentary. Corrections for multiple comparisons are used when several groups are studied at the same time, with the ANOVA test. When we studied anterior/posterior concha bullosa and lateral/medial superior attachment of the uncinate process we only performed one comparison with the chi2 test." Trying to educate me about statistics when the authors clearly have no idea what they are talking about is simply outrageous. Please carefully read up on multiple comparison correction or ideally, take an extensive statistics class. This quote is from wikipedia to keep it simple: "For example, if one test is performed at the 5% level and the corresponding null hypothesis is true, there is only a 5% chance of incorrectly rejecting the null hypothesis. However, if 100 tests are each conducted at the 5% level and all corresponding null hypotheses are true, the expected number of incorrect rejections (also known as false positives or Type I errors) is 5. If the tests are statistically independent from each other, the probability of at least one incorrect rejection is approximately 99.4%." So just to make this an educational review: No, multiple comparison correction is not only applied to ANOVA tests. If you run one ANOVA and it tells you that the groups are different, that is your answer and you do not need a multiple comparison correction. Only when you then run multiple t-tests or similar comparisons to find out which groups are different from each other, you will have to correct. The same is true here. The more tests you run (yes, also chi squared tests), the likelier it is to get a statistically significant result. With an alpha level of 0.05 and at least 8 comparisons whose p values are presented in table 1, you can type these variables in on a website like this (https://www.statology.org/bonferroni-correction-calculator/) and find out that - when using Bonferroni - you need a p that is smaller than 0.00625 to be statistically significant - which your results do not have! The number even comes with an interpretation, which I cite here: "If you conduct 8 comparisons, only reject the null hypothesis of each comparison if it has a p-value less than 0.00625." I hope you have now understood the problem, this is as basic as it gets when it comes to statistics. 

Thank you. Please, do not take our commentary to the personal. We do not know the reviewer, neither their knowledge in the field. Therefore, our commentaries were simple a friendly discussion.

We understand that there is a valid discussion here, even between statisticians. We have consulted with a professor from our university. What the reviewer has pointed out is true, of course. However, our point here is that it should be applied for an overall null hypothesis. For example, if we perform a comparison between cohorts to check the comparability between samples, should we include these comparisons in the final p adjustment?

For example, we can offer several relevant papers published in Q1 otolaryngology journals were p-value was not adjusted for multiple comparisons when the overall null hypothesis was different: DOI: 10.1016/j.jaci.2020.05.032 ; 10.1016/j.jaci.2017.05.044.

Even with this interesting discussion (that again, please, do not take to the personal), we have included a paragraph in the discussion (page 6, line 3-5)

Yes, it is the second largest group ever according to table 2, but what does it help if it doesn't offer new insights, adds 8 samples compared to the 2006 Ercan study, a paper published 16 (!) years ago, and doesn't even have half as many as Mahmutoğlu in 2015? But the most problematic thing is that the statistics are simply wrong (see above), which is why the paper cannot be fixed and you need a way larger sample to draw your conclusions.

Ok, we understand the point. Our response was not about if the paper offers or not new insights, it was about the commentary of the reviewer complaining about the sample size.

Regarding the other discussion, questioning if the paper offers or not new insights. If the initial question is valid, even papers with negative results should be published. If only papers with positive findings are published, then we are also missing important information, and other researchers could be conducting the same study as no results are available.

Author Response File: Author Response.pdf