Review Reports

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

This is a well-written article with an original question which in turn opens the door to other questions. I thank the authors for this very interesting reading.

I only have some thoughts that I would like to address. The duration of disease could have been an interesting feature in Table 1 and for the analysis, if available. In Methods, can you please combine the first two sentences of the paragraph into a single sentence (e.g.: “this observational cross-sectional study”) ?

Author Response

Reviewer #1

This is a well-written article with an original question which in turn opens the door to other questions. I thank the authors for this very interesting reading. I only have some thoughts that I would like to address.

The duration of disease could have been an interesting feature in Table 1 and for the analysis, if available.

Response:

We thank the reviewer for their suggestion. Unfortunately, disease duration was not collected from participants during the study. We tried to collect it retrospectively, but the data was deemed unreliable.

In Methods, can you please combine the first two sentences of the paragraph into a single sentence (e.g.: “this observational cross-sectional study”) ?

Response:

The manuscript has been amended according to the suggestion.

(lane 190, marked document): “This observational, cross-sectional study included patients with gout, during either flares or intercritical periods, seen at the rheumatology section of a tertiary academic hospital.”

Reviewer 2 Report

Comments and Suggestions for Authors

I have read the manuscript entitled "Length of Monosodium Urate Crystals in Synovial Fluid Based on Ultrasound Articular Deposits: Advancements in Crystallization Process."

Although the topic is interesting, several methodological issues, and above all, an overinterpretation of the results, make it difficult to draw firm conclusions from this work.

• The objective of the study is not clear: “we aimed to determine whether the length of MSU crystals in SF samples would vary according to organized crystal deposits in the joint (likely long crystals deriving from them).” If I understand correctly, the objective should rather be reformulated as: “the aim is to assess whether the length of the crystals is associated with the presence and characteristics of sodium urate deposits as detected by ultrasound,” is that correct?

• Reproducibility should not be assessed on the ability to differentiate between gout and CPPD crystals, but rather on the measurement of crystal length itself. Did you conduct such a reproducibility study?

• On page 13, you mention: “Our cumulative data suggest that MSU crystallization in SF begins…”, but you provide no original data from your study to support the statements that follow this sentence.

• Your study does not show any association between crystal length and clinical or ultrasound findings (except for one isolated variable: long crystals–P66). Yet you conclude: “In summary, MSU crystal length and the number of long crystals appear to be associated with organized crystal deposits identified by ultrasound (namely, DC signs and tophi).” Multiple statistical tests were performed, but your statistical methods do not account for multiple testing or adjust the p-value/significance threshold, exposing your findings to an increased risk of type I error (false positives).

Author Response

Reviewer #2

I have read the manuscript entitled "Length of Monosodium Urate Crystals in Synovial Fluid Based on Ultrasound Articular Deposits: Advancements in Crystallization Process." Although the topic is interesting, several methodological issues, and above all, an overinterpretation of the results, make it difficult to draw firm conclusions from this work.

Response:

We thank the reviewer for the time spent revising our manuscript. We will consider and answer the issues he/she raised individually.

The objective of the study is not clear: “we aimed to determine whether the length of MSU crystals in SF samples would vary according to organized crystal deposits in the joint (likely long crystals deriving from them).” If I understand correctly, the objective should rather be reformulated as: “the aim is to assess whether the length of the crystals is associated with the presence and characteristics of sodium urate deposits as detected by ultrasound,” is that correct?

This is a valuable contribution. The aim of the study should be clear and easily understood. We have amended it accordingly.

(lane 182, marked document): “To add more data that support this hypothesis, we aimed to assess whether the length of MSU crystals and the presence of long crystals are associated with the presence and characteristics of urate deposits as detected by ultrasound in the index joint.”

Reproducibility should not be assessed on the ability to differentiate between gout and CPPD crystals, but rather on the measurement of crystal length itself. Did you conduct such a reproducibility study?

This issue was raised by the editor as well. One major concern was that the two observers (a medical student and a rheumatology trainee) could correctly identify MSU crystals. Some artifacts, dust, etc., might be taken as MSU crystals, biasing our data (as some of them are often quite large). For this reason, we designed a brief, dedicated training on PLM and performed an interrater reliability assessment on crystal identification. The length measuring process in the display is quite simple for a needle-shaped crystal, measuring from end to end, so no formal reliability was tested. We think the repercussions on the study results are minimal. Still, we have disclosed in the Discussion as a limitation for the reader’s consideration.

(lane 650): “We acknowledge that no formal comparison was tested for length measurement, as we used a simple procedure.”

On page 13, you mention: “Our cumulative data suggest that MSU crystallization in SF begins…”, but you provide no original data from your study to support the statements that follow this sentence.

Thanks for the comment. Following the editor’s suggestion, we have extensively rewritten the Discussion. That statement has been modified.

(lane 496): “Available data suggest that MSU crystallization in SF begins with…”

Your study does not show any association between crystal length and clinical or ultrasound findings (except for one isolated variable: long crystals–P66). Yet you conclude: “In summary, MSU crystal length and the number of long crystals appear to be associated with organized crystal deposits identified by ultrasound (namely, DC signs and tophi).” Multiple statistical tests were performed, but your statistical methods do not account for multiple testing or adjust the p-value/significance threshold, exposing your findings to an increased risk of type I error (false positives).

This issue was also discussed with the editor. Accordingly, we have removed all references in the text indicating trends to significance. The current text now shows our findings in terms of raw data and statistical analysis. The Discussion and conclusion have been modified accordingly. Some examples of the modifications introduced in the Manuscript:

(lane 384): “A wider dispersion is observed in those with ultrasound deposits, but the difference was not statistically significant (p=0.42).”

(lane 414): “Crystal length in joints that met model 2 criteria was similar to those without organized deposits (22.78µm vs. 21.19µm, p=0.12) (Figure 2).”

(lane 418): “The difference in long crystals classified using the P75 and P90 definitions did not reach statistical significance.”

We did not adjust the p-value despite multiple testing, and we acknowledge being at risk of type I error. However, our findings are consistent with the raw data presented in Figures 2-4. There, it is evident that the numbers suggest a higher prevalence of long crystals in cases with ultrasound deposits, although statistical significance was only reached for the P66 definition. Having that for both the absolute and relative number of P66 crystals is also noteworthy. Our data also suggest longer crystals using P75 and P90 definitions, but as stated above, the difference was not significant, and this is how it is presented in the manuscript.

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

The authors have answered to the the comments and questions but methodological flaws persist

Author Response

Comment: "The authors have answered to the the comments and questions but methodological flaws persist"

Reply: Thank you for your thoughtful review, which has undoubtedly improved our manuscript.