The Use of Diffusion Weighted Imaging and Intravoxel Incoherent Motion to Assess Edema and Perfusion in Abdominal Organs during Endotoxemia
Round 1
Reviewer 1 Report (Previous Reviewer 2)
This study evaluated the usefulness of DWI/ADC/f value in evaluating the perfusion and edema of tissues using the piglet model with endotoxin infusion. The results of this study as a basic research may be valuable for further research on MRI evaluation.
Author Response
We want to thank the reviewer for accepting the manuscript without any changes needed.
We are glad that the paper was found interesting and feasible for publication.
Reviewer 2 Report (New Reviewer)
General Comments
The main aim of the paper is to assess whether Diffusion can be a suitable technique to assess edema and perfusion in abdominal organs (liver, spleen, intestine, and kidneys) in a porcine model of endotoxemia.
The same authors have already published a paper on the same topic in 2018.
Marchesi, S., Ortiz Nieto, F., Ahlgren, K. M., Roneus, A., Feinstein, R., Lipcsey, M., ... & Hedenstierna, G. (2019). Abdominal organ perfusion and inflammation in experimental sepsis: a magnetic resonance imaging study. American Journal of Physiology-Gastrointestinal and Liver Physiology, 316(1), G187-G196.
Since the previous paper uses the same experimental model and, basically, the same methods, I don’t think the current paper has enough innovation to be considered for publication. The inclusion of a further group identified as LTD (normal hemodynamic and increased abdominal edema) does not justify the need for a new paper.
Furthermore, the methodological approach, especially regarding image processing and analysis, is not well described and this is a critical aspect to evaluate the quality of the presented results.
The results, especially those regarding MRI, are not well reported since no relevant images are included in the paper.
The overall structure of the paper doesn’t seem to fulfill the journal guidelines. Indeed Materials and methods are described after the discussion and before the conclusions. The typical outline for this journal is as follows: Introduction, Materials and Methods, Results, Discussion, and Conclusions.
English language must be thoroughly revised in the document.
Some specific comments
Title
I would suggest the authors refer to both IVIM and DWI
Abstract
IVIM is a technique that can be used to probe perfusion in biological tissues. Of course, IVIM is a particular model of diffusion MRI and I would suggest referring to IVIM as well as DWI within the title and abstract.
I would change the term f value with perfusion fraction since it is more correct and generally accepted.
How many groups were considered in the study? 3 or 4? In the materials and methods section, 4 groups are mentioned.
Some methodological details about the MRI protocol must be included in the abstract.
Line 29. Please change “is” with “in”
Introduction
Page 2 line 47. Could you please cite some reference articles to justify the sentence?
Page 2 lines 47-48 even in this case please add references. Are you referring to cytotoxic edema?
Page 2 lines 71-73. There’s a review in which “normal” values of IVIM parameters are compared to those in pathological conditions in the liver.
Li, Y. T., Cercueil, J. P., Yuan, J., Chen, W., Loffroy, R., & Wáng, Y. X. J. (2017). Liver intravoxel incoherent motion (IVIM) magnetic resonance imaging: a comprehensive review of published data on normal values and applications for fibrosis and tumor evaluation. Quantitative imaging in medicine and surgery, 7(1), 59.
The previous paper by the same authors is not mentioned.
Taking into consideration the scientific literature, the rationale of the paper is questionable and the topic is not novel.
Methods
The methodology related to image processing and analysis is poorly described. For example, which algorithms did the authors use for data fitting? Bayesian methods or even better Deep Learning approaches have shown better performances if compared to LSQ methods. The authors must consider this for their analysis. Which kind of pre-processing was performed on the images? Any registration or filtering?
Why parametric methods were used in the statistical analysis? Did the authors check the “normality” of their data? If not, I would suggest doing so and, in case, using just non-parametric tests.
Results
I do not see any relevant plate showing MR data that can help the reader to see differences among groups. This is mandatory.
Author Response
We want to thank the reviewer for the comments and the through analysis of our paper.
We think the importance of this paper compared to the previous one, is not only to have add one more animal group but to give more information about the MRI results, describing them more throughly. It is important for the authors to collect the information coming from the images analysis in one paper, focusing only on them (adding for example the analysis of D). In our opinion, this is innovative enough and the amount of information added is of great importance.
The paper was reviewed by a native english speaker.
Specific comment
Title: we add also IVIM to the title.
Abstract: we addressed all the reviewer's suggestions. Unfortunately the max limit of words in the abstract makes it impossible to add technical details.
Introduction: we addressed all the reviewer's suggestions. We politely disagree with the reviewer on the non-novelty of the topic. There are no studies performing MRI in sepsis to evaluate edema and perfusion of abdominal organs, and our opinion is that this is novel enough to write a paper about our findings.
We added the reference the reviewer asked for.
Methods: we add the required info in the new manuscript version. We tested the data normality. We add a sentence on the manuscript.
Results: we didn't understand the reviewer comment. Is the reviewer saying we didn't present differences among the groups in a clear way? We think that Figure 2 and 4 present the results on perfusion fraction and edema in all the groups including the differences among the groups in a clear way. If the reviewer refers to something different, we please them to be more specific and we will be happy to address their concerns.
Reviewer 3 Report (New Reviewer)
Rather interesting paper. The MRI methods description and the presented results are incomplete. Poor abstract. Too much method description in the Results section.
Specific comments: 1. the abstract is unclear, especially that the phenomenon perfusion and anatomy (edema) are presented as similar properties. 2. It's incorrect to state that DWI is not clinical standard for tissues other than brain. 3. abstract line 19: of them>pigs. 4. Line 29: is>in. 5. line 107: ROI sizes must be decribed in M&M, not here. 6. line 110: Unclear sentence also belonging to M&M. What do the ROI fits<95% mean? Probably refers to the fitting of D*, D and f from the signal intensities of ROIs on the subsequent DWI images. Please be clear. 6. Table 1 contains unexplained abbreviations. 7. ADC values are reported, but sorely missing is information from what b value images they were calculated. 8. It is much more interesting to report the D values, refecting diffusion, rather than ADC reflecting a b value dependent mixture of perfusion and diffusion. Table 2: MRI details are lacking, such as the pixel size and the fat suppression used. 9. Fig.6 can not be assessed: too small and unsharp, so the procedures and formulas can not be checked.
Author Response
We want to thank the reviewer for a through analysis of our manuscript.
Following, our answer to the reviewer's questions/concerns.
- We struggle understanding what the reviewer means. We do not consider perfusion and edema as similar properties. We also do not think that edema is related to the anatomy of the tissue. Our definition of perfusion is the passage of blood in the capillary system which is related to the oxygen delivery to tissues. And for edema we intended extravascular water related to the increase permeability of the capillaries induced by the septic state. We have added definitions in the introduction.
2. We didn't state DWI is not used outside the brain, on the contrary, in the Introduction we described how the technique has been employed successfully outside the neurological field, also in abdominal organs. The novelty of the presented study is to use DWI to assess edema and perfusion during sepsis, which is not a clinical standard. To the authors knowledge there are no other papers that investigate a similar use of the technique.
3. Done
4. Done
5. We discuss the positioning of this sentence. We think that this sentence belongs to the results section as the ROIs size is not decided in advance and part of the protocol. We tried to have homogeneous ROIs size through the groups but that depended on the anatomy of the different animals. So we consider it as part of the results and not the methods.
6. The ROIs had to have a goodness of fit higher than 95%.
6. Abbreviations added
7. B values are reported in Table 2.
8. D results are reported in fig. 4 and considered throughout the entire paper. If D is not always used as a comparison to other studies in the discussion is only because to find literature where D is measured in organs of healthy patients was difficult. Is there any specific suggestions of the reviewer to better highlight D results?
Table 2 was updated
9. Figure 1 has been upload separately in higher definition.
Author Response File: 
Author Response.pdf
Round 2
Reviewer 2 Report (New Reviewer)
I'm still convinced that the paper is very much similar to the previous published by the same group and that the major novelty was already shown in that paper.
Anyway, the authors must declare in their manuscript that the current paper is an incremental work based on the previous one. This aspect must be stressed in the paper (e.g, in Introduction and Discussion).
Furthermore, the methodology the authors used for data fitting is not the best solution for IVIM data analysis.
Segmented Least Squares approaches, or better Bayesian methods or, even better, Deep Learning approaches can be much more reliable. This aspect must be at least discussed and some papers must be cited:
- While, P. T. (2017). A comparative simulation study of bayesian fitting approaches to intravoxel incoherent motion modeling in diffusion‐weighted MRI. Magnetic resonance in medicine, 78(6), 2373-2387.
- Gustafsson, Oscar et al. Impact of prior distributions and central tendency measures on Bayesian intravoxel incoherent motion model fitting. Magn Reson Med . 2018;79:1674–1683
- Orton, M. R., Collins, D. J., Koh, D. M., & Leach, M. O. (2014). Improved intravoxel incoherent motion analysis of diffusion weighted imaging by data driven Bayesian modeling. Magnetic resonance in medicine, 71(1), 411-420.
- Lanzarone, E., Mastropietro, A., Scalco, E., Vidiri, A., & Rizzo, G. (2020). A novel bayesian approach with conditional autoregressive specification for intravoxel incoherent motion diffusion‐weighted MRI. NMR in Biomedicine, 33(3), e4201.
- Kaandorp, M. P., Barbieri, S., Klaassen, R., van Laarhoven, H. W., Crezee, H., While, P. T., ... & Gurney‐Champion, O. J. (2021). Improved unsupervised physics‐informed deep learning for intravoxel incoherent motion modeling and evaluation in pancreatic cancer patients. Magnetic resonance in medicine, 86(4), 2250-2265.
- Mastropietro, A., Procissi, D., Scalco, E., Rizzo, G., & Bertolino, N. (2022). A supervised deep neural network approach with standardized targets for enhanced accuracy of IVIM parameter estimation from multi‐SNR images. NMR in Biomedicine, 35(10), e4774.
Author Response
We thank the reviewer for giving us the opportunity to improve our work.
We better described our previous work at the end of the introduction, as asked by the reviewer.
Besides, we cited the alternative methods for images analysis as a limitation in the dedicated paragraph.
Best wishes
The authors
Reviewer 3 Report (New Reviewer)
Acceptable revision.
Author Response
We want thank the reviewer for accepting our revision.
Best wishes
The authors
Round 3
Reviewer 2 Report (New Reviewer)
I want to thank the authors for their improvements. However, some minor changes are still needed.
- The previous paper at the end of the introduction (lines 92-94) must be cited. In the current version the citation is missing.
- Discussion (from line 320). The authors must clearly explain that the fitting approach they used can introduce bias especially at low SNRs and for the estimation of f and D* maps.
Author Response
We want to thank the reviewer for the comments.
We changed the manuscript according to their advices.
We hope that the reviewer can now consider the manuscript ready for publication.
This manuscript is a resubmission of an earlier submission. The following is a list of the peer review reports and author responses from that submission.
Round 1
Reviewer 1 Report
I accepted to review this manuscript because I possess expertise in NMR spectroscopy. The fact that the manuscript was submitted to the journal Magnetochemistry made me think that such expertise will be needed to assess the quality of the manuscript. Having read now the manuscript, the subject is much more on medicine rather than on magnetic resonance. Furthermore, there is definitely no connection to chemistry as well. This journal does not seem to be the suitable home for this manuscript.
I am not an expert in medicine, and thus unable to assess the validity of the medical procedures and analyses undertaking in the presented study. That said, the authors made it extremely difficult for the reader to follow the connection they are trying to make between diffusion measurements and analytical value for some medical assessment. It does not help that the manuscript does not define all terms and acronyms, poorly lays out the experimental details, and is riddled with imprecise language and sloppiness. For here, I am bringing up just a couple examples to illustrate the point: The very first sentence of the abstract is incomprehensible, and the Figure labels are not chronological. The authors need to explain more carefully their initial thinking why self-diffusion measurements might be a (quantitative?) measure of endema and perfusion, how this thinking led to the design of the study, and in how far the results support or not support their initial ideas. In addition, the results are more or less just listed without helping the reader to see their significance in how far these fit into the bigger picture. Thus, even if the subject of the manuscript would be suitable for another MDPI journal, it needs major revisions.
Reviewer 2 Report
This study evaluated the usefulness of DWI/ADC/f value in evaluating the perfusion and edema of tissues using the piglet model with endotoxin infusion. The results of this study as a basic research may be valuable for further research on MRI evaluation. Following are the concerns before the publication.
Line 328
The description that piglets were used should also be added to the Materials and Methods section.
Line 182 Figure 4
Why the results of Kidney were omitted in Figure 4? In other Figures (3, 5) the results of kidney were shown.
Line 209 Figure 6
Were there any co-relationship between ADC, D and f-value? In addition, the scatter plot of urinary output (perfusion) and ADC/D was shown whereas the scatter plot of f-value is not shown. The overall scatter plot between f-value and Wet/Dry (edema) can be shown if there were any clear relationship.
The results of Lactate and SvO2 were shown to check the validity of the endotoxin-induced septic model. Was there any direct relationship between the ADC/D/f-value and these laboratory data?
