Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Dear author 

Congratulation for your hard and excellent work. However, I have some questions listed  below. 

1. Please clarify if the "insufficient group" suggest vit D level 10-30 ng/mL, and "deficiency group suggest vit D <10 ng/mL. Current description in manuscript may suggest the deficiency group is including in the insufficient group. 

2.  Are the p values in table 1 and table 2 are results of two total group ? Did you make any subgroup analysis ?

I suggest that may consider compare the percentage of severe disease between 2 vit D level groups, or compare the scores of SLEDAI. That is to make some subgroup analysis may help identify the importance of vit D level. 

Author Response

1. Please clarify if the "insufficient group" suggest vit D level 10-30 ng/mL, and "deficiency group suggest vit D <10 ng/mL. Current description in manuscript may suggest the deficiency group is including in the insufficient group. 
   1. Thank you for your comment. For the group of patients with vitamin D deficiency, levels between 10 and 29 ng/mL were considered (so less than 30 ng/mL and up to 10 ng/mL); and for patients with deficiency, those with <10 ng/mL were considered (so patients with levels starting at 9 ng/mL). We have changed the wording in the Methods section (and elsewhere) to more clearly describe the levels used to divide the groups.
2. Are the p values in table 1 and table 2 are results of two total group? Did you make any subgroup analysis? I suggest that may consider compare the percentage of severe disease between 2 vit D level groups, or compare the scores of SLEDAI. That is to make some subgroup analysis may help identify the importance of vit D level. 
   1. The results presented in Tables 1 and 2 are the results of the inferential analysis of the comparison between the two groups (deficient vs. insufficient); in cases where dichotomous categorical variables were analysed, the chi-squared test was used in a 2x2 contingency table (thus comparing proportions), while for the other variables a Student's t-test was performed. Regarding subgroup analysis, due to the small sample size we decided not to perform this type of analysis (which would result in comparisons of very small groups and of low statistical power), however, following your suggestion, we have included the results of an analysis comparing total SLEDAI-2K scores between the two groups, as well as the median SLEDAI-2K score of those classified as low-moderate and severe activity between the deficient and insufficient groups (this information has been added to the text in the results section), and a small paragraph in the discussion section where we discuss these new results.
      1. Additional text Results: Patients with vitamin D deficiency had a median SLEDAI-2K score of 12 (interquartile range (IQR)=11), while patients with insufficiency had a median score of 9 (IQR=13), which was not a significant difference (p=0.22). Patients with vitamin D deficiency, both in the group classified as low-moderate SLEDAI-2K score and in the group classified as severe, had a significantly higher median total score than patients with vitamin D insufficiency, with a median low-moderate SLEDAI-2K score=11 (IQR=4) vs. 7 (IQR=4), p=0.003, and a median severe SLEDAI-2K score=26 (IQR=12) vs. 20 (IQR=4), p=0.03.
      2. Text discussion added: We also found no significant differences between the total SLEDAI-2K scores between the two groups, we only found that in the group of patients with vitamin D deficiency, the total SLEDAI-2K scores were significantly higher when divided into the group with low-moderate scores and the severe group, compared with the patients with vitamin D insufficiency, which may indicate that with a larger sample size, we may find significant differences related to disease activity.

Reviewer 2 Report

Comments and Suggestions for Authors

In recent decades, the treatment of systemic lupus erythematosus (SLE) has shifted	55
from the use of hydroxychloroquine, systemic glucocorticosteroids, and conventional im-	56
munosuppressive drugs to the use of biological agents [1].

Comment: If hycroxychloroquine etc. are still used, the verb should be “tended to shift”

 

vitamin D levels

25-hydroxyvitamin D

Comment: Only one term should be used in the manuscript. 25-hydroxyvitamin D concentration is more correct than vitamin D level. If 25-hydroxyvitamin D is to be used, it should be 25-hydroxyvitamin D [25(OH)D] the first time in both the abstract and text, then 25(OH)D thereafter.

 

In the data reported in ENSANUT in 2006 [25], it was found that	95
the average serum level of vitamin D in Mexican adults over 20 years of age was 38.8	96
ng/ml,

Comment: This value seems to be very high. Please check the assay used and what is known about the accuracy of that assay.

 

Serum vitamin D levels were measured using a chemiluminescence immunoassay for the 143 quantitative determination of 25-hydroxyvitamin D. In the first incubation step, 25-OH 144 vitamin D was dissociated from its binding protein and bound to a specific antibody on a 145 solid phase. After 10 minutes, a tracer (vitamin D conjugated to an isoluminol derivative) 146 was added, followed by a second incubation for 10 minutes. Unbound material was re- 147 moved during a wash cycle, and initiator reagents were added to trigger the chemilumi- 148 nescent reaction.

Comment. This discussion should include the name of the instrument, the manufacturer, city (state) and country.

Also, discuss the precision , accuracy and how the system was calibrated and what is known about how it compares with standardized 25OHD values.

Also, since 25(OH)D was measured anytime during the year, seasonal variations may have occurred. Suggest presenting a plot of 25(OH)D vs. month or season of the year and discussing whether seasonality could explain any of the findings.

 

Footnotes to tables. The acronyms should be listed in alphabetical order.

 

Keywords: Also in alphabetical order.

 

Figure 1. Suggest looking at the three patients who had high 24-hour protein to see what may have caused that. Could a protein-rich diet around that time explain it?

 

References: I suggest searching Google Scholar with this search string: “systemic lupus erythematosus vitamin D”, starting with papers published from 2020 to present.

Here are some of the findings:

Vitamin D and systemic lupus erythematosus: Causality and association with disease activity and therapeutics

LJ Ho, CH Wu, SF Luo, JH Lai - Biochemical Pharmacology, 2024 - Elsevier

Vitamin D supplementation in systemic lupus erythematosus: relationship to disease activity, fatigue and the interferon signature gene expression

R Magro, C Saliba, L Camilleri, C Scerri, AA Borg - BMC rheumatology, 2021 - Springer

 

 

Significant digits. The general rule is that no more non-zero digits should be given than are justified by the uncertainty of the value.

See "Too many digits: the presentation of numerical data"

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4483789/

                                   

If the uncertainty (or difference when comparing numbers) is greater than about 7%, only two non-zero digits are justified.

P values should be given to two decimal places unless the first two are 00 or the number lies between 0.045 and 0.054. If the first two are 00, then only one non-zero digit can be given.

 

Thus, p values should be adjusted

Percentages should be given in whole numbers

Numbers such as these should be revised to have fewer non-zero digits:

24-hour urine protein (1.801 vs. 1.0569 g/24 should be 24-hour urine protein (1.8 vs. 1.1 g/24

emoglobin (g/dl)

9.90

3.95

10.34

5.30

Should be

emoglobin (g/dl)

9.9

4.0

10.3

5.30

 

Platelets (x109/L)

128

125.35

196.50

228.57

Should be

Platelets (x109/L)

128

125

197

228.57

 

Table 3

24-hour urine protein

-0.0711

0.005

0.6164

Should be

24-hour urine protein

-0.07

0.005

0.62

Please review all numbers in abstract, text, tables, and figures and adjust accordingly.

Author Response

1. Comment: If hycroxychloroquine etc. are still used, the verb should be “tended to shift”
   1. The text has been changed to reflect your suggestion.
2. Comment: Only one term should be used in the manuscript. 25-hydroxyvitamin D concentration is more correct than vitamin D level. If 25-hydroxyvitamin D is to be used, it should be 25-hydroxyvitamin D [25(OH)D] the first time in both the abstract and text, then 25(OH)D thereafter.
   1. We have made the change at your suggestion.
3. Comment: This value seems to be very high. Please check the assay used and what is known about the accuracy of that assay.
   38. In all the references consulted and cited in the article, as well as in the international literature, we did not find any information describing the levels of 25-(OH)D reported in ENSANUT 2006 (38.8 ng/mL) as particularly high; they are levels that are considered normal (considering that levels above 150 ng/mL are considered toxic, as mentioned in reference 26 of our article, which refers to the article by Holick) and, as described in our article, they are average values for Mexican adults over 20 years of age. There are no data on average levels in SLE patients on a large scale in Mexico, only data from the study by García-Carrasco (which we cite (reference 17)) and which we briefly describe in the article. At ENSANUT 2006, the analysis of 25-OH-D3 in biological serum samples was performed by direct ELISA (25(OH)-Vitamin D direct ELISA Kit), manufactured by the German company Immunodiagnostik AG and marketed by the North American company ALPCO Immunoassays. The method uses an anti-25(OH)-vitamin D monoclonal antibody with a recognition specificity for 25(OH)-vitamin D3 of 100%, for 25(OH)-vitamin D2 of 67.8%, for 24,25(OH)-vitamin D3 ≥ 100% and for vitamin D2 (ergocalciferol) of 0.3%. The intra- and inter-assay variation was less than 10%. This type of analysis is generally performed in most clinical laboratories and is methodologically similar to that reported in the various studies of 25-(OH)D. Most of the studies known to us used either ELISA or chemiluminescence (as in our study).
4. This discussion should include the name of the instrument, the manufacturer, city (state) and country.
   1. Following your comment, we have added the data from the chemiluminescence kit used to obtain the results used in the study.
5. Also, discuss the precision , accuracy and how the system was calibrated and what is known about how it compares with standardized 25OHD values.
   1. One of the clarifications that we have to make and the reason why data on the calibration, precision and sensitivity of the method used to measure 25-(OH)D are not included in our article is that our study is a retrospective study and therefore the method of measuring 25-(OH)D was not chosen for this study, but the results of the measurement were taken from the records and the methodology was requested from the central hospital laboratory. However, we have added information about the calibration (which was carried out according to the manufacturer's recommendations and supervised by the quality control department of the hospital's clinical laboratory) and we have added to the limitations section that, although the method used to measure 25-(OH)D is a precise method and even more sensitive than the ELISA, it is not the method used in most of the articles, but we have included two references that demonstrate the equivalence of the results.
      1. New text in methods: Calibration was performed according to the manufacturer's recommendations and supervised by the quality control department of the hospital's clinical laboratory.
      2. New text in the limitations: Another limitation to consider is that although the method used to measure 25-(OH)D is a precise method and even more sensitive than the ELISA, it is not the method used in most articles. However, there are several articles reporting the equivalence of results of different variables measured in serum, including 25-(OH)D [45,46].

• New reference 45: Yun C, Chen J, Yang C, Piao J, Yang X. [Comparison for ELISA and CLIA of serum 25-hydroxy vitamin D determination]. Wei Sheng Yan Jiu. 2015 May;44(3):435-9. Chinese.

12. New reference 46: Enko D, Kriegshäuser G, Stolba R, Worf E, Halwachs-Baumann G. Method evaluation study of a new generation of vitamin D assays. Biochem Med (Zagreb). 2015 Jun 5;25(2):203-12. doi: 10.11613/BM.2015.020.

6. Also, since 25(OH)D was measured anytime during the year, seasonal variations may have occurred. Suggest presenting a plot of 25(OH)D vs. month or season of the year and discussing whether seasonality could explain any of the findings.
   1. Unfortunately, we did not collect information on the date of 25-(OH)D measurement in the database. Although we agree that there are reports of seasonal variation in 25-(OH)D levels (which we mentioned as one of the possible explanations for the results obtained), due to the small sample size obtained, we could not provide sufficient data to be able to report 25-(OH)D levels of patients in a similar season of the year (so it was not considered for this exploratory study). Although we have already mentioned this in the discussion, we have decided to add to the limitations the corresponding information about possible seasonal variations in 25-(OH)D levels. Even with this limitation, we believe that the data obtained are relevant because of the limited information on this subject in the Mexican population and the exploratory nature of this study, which aims to allow future prospective controlled studies in which a greater number of variables can be controlled to allow greater statistical certainty.
      1. New text added in the limitations: Finally, in this exploratory study, we did not take into account the seasonal variation in serum 25-(OH)D levels [26], which could affect the results obtained, but we believe that the data obtained are relevant due to the limited information on this topic in the Mexican population and the exploratory nature of this study, which aims to allow future prospective controlled studies in which a greater number of variables can be controlled to allow greater statistical certainty.
   2. Footnotes to tables. The acronyms should be listed in alphabetical order.
      1. We have changed the order of the abbreviations as you suggested.
   3. Keywords: Also in alphabetical order.
      1. We have made the suggested changes.
   4. Figure 1. Suggest looking at the three patients who had high 24-hour protein to see what may have caused that. Could a protein-rich diet around that time explain it?
      1. The protein levels reported in these three patients are higher than 3 g in 24 hours (this cannot be caused by dietary reasons alone), which is indicative of nephrotic syndrome and therefore may be associated with target organ damage caused by SLE and an indicator of high disease activity. For this reason, in the discussion we are inclined to follow the interpretation of Athanassiou et al. and interpret this association as an indicator of the relationship between 25-(OH)D and SLE activity level, taking into account the limitations of our study and the need for future confirmation of this result.
   5. References: I suggest searching Google Scholar with this search string: “systemic lupus erythematosus vitamin D”, starting with papers published from 2020 to present.
      1. For our article, we considered several articles published after 2020 as recent references to research on 25-(OH)D and SLE. However, based on your recommendation, we added some data that we found interesting and relevant from the two articles suggested in the discussion section. However, neither of the two articles has substantial information that adds anything different to what we present with the cited references and, in addition, much of the recent research (including these two suggested articles) has focused on evaluating the usefulness of vitamin D supplementation in patients with SLE in populations where the high prevalence of low 25(OH)D levels in patients with SLE has been widely studied, which is not the case in the Mexican population, so we feel that our exploratory study adds relevant information to justify future research on the relationship between 25(OH)D and SLE.
         1. New text added to the discussion: Although the evidence for the benefit of vitamin D supplementation in patients with 25-(OH)D deficiency or insufficiency is mixed and sometimes conflicting, it has been suggested that, because of the frequent and prolonged use of corticosteroids in patients with SLE and the high prevalence of low levels of 25-(OH)D in this population, supplementation may be particularly beneficial and may help to reduce the doses of corticosteroids used in these patients [47,48].
         2. New reference 47: Ho LJ, Wu CH, Luo SF, Lai JH. Vitamin D and systemic lupus erythematosus: Causality and association with disease activity and therapeutics. Biochem Pharmacol. 2024 Sep;227:116417

• New reference 48: Magro R, Saliba C, Camilleri L, Scerri C, Borg AA. Vitamin D supplementation in systemic lupus erythematosus: relationship to disease activity, fatigue and the interferon signature gene expression. BMC Rheumatol. 2021 Dec 3;5(1):53

11. Significant digits. The general rule is that no more non-zero digits should be given than are justified by the uncertainty of the value. If the uncertainty (or difference when comparing numbers) is greater than about 7%, only two non-zero digits are justified. P values should be given to two decimal places unless the first two are 00 or the number lies between 0.045 and 0.054. If the first two are 00, then only one non-zero digit can be given. Thus, p values should be adjusted.
    1. We have made the suggested changes throughout the article.
12. Percentages should be given in whole numbers
    1. We have made the suggested changes throughout the article.
13. Numbers such as these (tables and figures) should be revised to have fewer non-zero digits. Please review all numbers in abstract, text, tables, and figures and adjust accordingly.
    1. We have made the suggested changes throughout the article.

Reviewer 3 Report

Comments and Suggestions for Authors

Dear authors,

You have mentioned that the Vit D defficiency is more frequent among younger patients. Could stratification by age lead to different statistical results?

Author Response

1. You have mentioned that the Vit D defficiency is more frequent among younger patients. Could stratification by age lead to different statistical results?
   1. We appreciate your comment. Due to the small number of patients included in our study, we felt that stratifying them by age group would give us very small groups of little value for analysis. However, based on your observation, we have added to the results the result of a linear regression analysis between age and vitamin D levels, which confirms what was reported in the bivariate analysis, obtaining a significant (p=0.03) and positive relationship (correlation coefficient=0.3), which means that in our sample of patients, the younger the age, the lower the serum 25-(OH)D levels (or that, the levels of 25-(OH)D increase with age).
      1. New text added to results: However, we found a significant (p=0.03) and positive relationship (correlation coefficient=0.3) between age and 25-(OH)D levels.
      2. New text added to the discussion: This result is consistent with what we found in the linear regression analysis between 25-(OH)D levels and age, where we found a significant (p=0.03) and positive relationship (correlation coefficient=0.3) between age and 25-(OH)D levels.

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

The authors had made their effort to correlate the vit D level and clinical significance. However, most of variables lack statistical significance. Maybe increase the patient number should help solve this difficulty. 

Author Response

The authors had made their effort to correlate the vit D level and clinical significance. However, most of variables lack statistical significance. Maybe increase the patient number should help solve this difficulty. 

Once again, we sincerely thank you for your thoughtful review and constructive comments on our manuscript. We acknowledge that most of the results did not reach statistical significance based on the conventional p-value threshold. However, in studies with small sample sizes, significant results—when they occur—can carry greater weight, as they suggest meaningful differences with a lower probability of arising by chance, even if these findings are limited to the specific cohort analyzed. We recognize that the retrospective nature of our study presents inherent limitations, including challenges in identifying all sample characteristics that might influence the outcomes. These factors, however, could plausibly relate to variations in 25-(OH)D levels. The limited sample size was primarily dictated by the availability of patients meeting the inclusion criteria, as well as the exploratory and retrospective scope of this investigation. Our primary objective was to bring attention to the potential relevance of variations in 25-(OH)D levels in patients with systemic lupus erythematosus (SLE), a common but understudied phenomenon in our population, as also noted in the global scientific literature. We fully agree that increasing the sample size would enhance the power to detect statistically significant differences. However, we would also like to emphasize the importance of interpreting p-values cautiously, as larger sample sizes can produce smaller p-values, which may not always reflect clinically meaningful differences. In light of your valuable feedback, we have further clarified the study’s limitations in the Results section to underscore the exploratory nature of our findings and the need for future research. Prospective studies with larger and more diverse populations are certainly warranted to build upon the observations we present here. Despite its limitations, we believe that our study offers a meaningful contribution by highlighting the importance of investigating factors associated with 25-(OH)D variations in Mexican patients with SLE, an area that remains underexplored.