Serum 25-Hydroxyvitamin D Concentration as a Biomarker and Immunomodulator in Patients with Acute Ischemic Stroke: A Retrospective Single-Center Study

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The article by Switonska and colleagues investigates the relationship between vitamin D status and the evolution of the AIS, comprehensively evaluating inflammatory biomarkers derived from basic laboratory parameters across vitamin D quartiles.

First, the references should be numbered in the order they appear in the text, so the authors should revise both the reference list and the citations in the manuscript.

Line 127: Please define the abbreviation HAI as the first occurrence in the text.

Also, evaluation of vitamin D levels according to TOAST classification of AIS may provide additional information on the relationship between vitamin D status and the underlying etiopathogenic mechanisms of acute ischemic stroke.

Since different TOAST subtypes are associated with distinct vascular and cardiac risk profiles, analyzing serum 25(OH)D concentrations across these categories could help determine whether vitamin D deficiency is more strongly associated with specific stroke mechanisms, particularly large-artery atherosclerosis or cardioembolic stroke.

Such an approach may improve the understanding of the potential role of vitamin D in stroke pathophysiology and prognosis, contributing to a better risk stratification and individualized management of patients with acute ischemic stroke.

The discussion section appears relatively short, considering the large number of evaluated parameters. It may be useful to provide a more detailed interpretation of the observed associations between vitamin D levels and inflammatory and immunometabolic indices, considering the existing literature. Since vitamin D exerts important immunomodulatory and anti-inflammatory effects, a more detailed discussion of these relationships could help clarify their potential pathophysiological and prognostic significance in acute ischemic stroke and further strengthen the clinical relevance of the study.

 

Author Response

1. The references should be numbered in the order they appear in the text, so the authors should revise both the reference list and the citations in the manuscript.

We reduced the number of references cited, and next we checked their citation in the text

2. Line 127 (now line 157): Please define the abbreviation HAI as the first occurrence in the text.

The abbreviation “HAI” was defined in the line 55 in first version of amnuscript, and in line 82 in revised version of manuscript.

3. Evaluation of vitamin D levels according to TOAST classification of AIS may provide additional information on the relationship between vitamin D status and the underlying etiopathogenic mechanisms of acute ischemic stroke. Since different TOAST subtypes are associated with distinct vascular and cardiac risk profiles, analyzing serum 25(OH)D concentrations across these categories could help determine whether vitamin D deficiency is more strongly associated with specific stroke mechanisms, particularly large-artery atherosclerosis or cardioembolic stroke. Such an approach may improve the understanding of the potential role of vitamin D in stroke pathophysiology and prognosis, contributing to a better risk stratification and individualized management of patients with acute ischemic stroke.

 

We analyzed vitamin D concentration in regard TOAST classification (line 162-168; line 249-256) in two ways. First we compared serum 25(OH)D concentration between four TOAST categories, but we failed to find any statistically significant differences. Moreover, we performed analysis of prevalence of respective TOAST categories in respective vitamin D quartiles. Although overall Chi² analysis in crosstabulation (multi-way table) showed significant relationships between TOAST categories and vitamin D quartiles, the differences in percentage of respective TOAST category between respective quartiles showed only single statistically significant associations.

[line 250-257] With regard to the TOAST classification of presumed AIS etiology, the study included 432 patients (31.3%) with LAA, 443 (32.0%) with CE, 172 (12.5%) with SVO, and 334 (24.2%) with SUE. Serum 25(OH)D concentration did not differ significantly between the TOAST categories (22.0 ± 15.8 vs. 20.8 ± 15.3 vs. 21.5 ± 14.4 vs. 21.5 ± 16.0 ng/ml, respectively; ANOVA, F = 0.433; p = 0.729). Moreover, although the overall chi-square (χ²) test showed a statistically significant association between TOAST category and 25(OH)D quartile, only a single statistically significant differences were found across the etiological stroke categories and 25(OH)D quartiles (Table 1).

4. The discussion section appears relatively short, considering the large number of evaluated parameters. It may be useful to provide a more detailed interpretation of the observed associations between vitamin D levels and inflammatory and immune-metabolic indices, considering the existing literature. Since vitamin D exerts important immunomodulatory and anti-inflammatory effects, a more detailed discussion of these relationships could help clarify their potential pathophysiological and prognostic significance in acute ischemic stroke and further strengthen the clinical relevance of the study.

 

We completed discussion with the section concerning  the relationships between serum vitamin D concentration and inflammatory indices, cardiovascular risk factors and comorbidities.

 

Line 353-372

Moreover, to the best of our knowledge, our study is the first to identify associations between serum 25(OH)D concentration and values of known and validated indices of inflammatory response to brain injury [3], such as the CRP, neutrophil-to-lymphocyte ratio, CRP-to-platelet ratio, SII, SIRI, and NPS (Table 2). This observation might suggest the real-world importance of 25(OH)D activity as a modulator of systemic, and probably cerebral, inflammatory response to AIS [3, 4, 35]. However, the power of the associations observed was weak and with low determination coefficient (R²) values. The relationships showed that serum 25(OH)D concentration determines only 1.7-3.6% of variances in the indices mentioned. This observation only weakly corroborates the results of a study in rats showing vitamin D as a strong immunomodulator mitigating the influence of brain ischemia/reperfusion-induced systemic and local inflammation [35, 45]. Nevertheless, we confirmed, known form literature [27], the importance of serum 25(OH)D concentration as an independent factor increasing the probability of a low disability score (mRS ≤ 2) at discharge on average by 52% for every quartile (Q1 à Q4), along with the neutrophil-to-lymphocyte ratio, a known and validated index of systemic inflammatory response to brain injury [3, 4], older age, diabetes mellitus, and treatment with EMT as a source of ischemia- reperfusion-induced neuroinflammation.

 

line 373-385

As mentioned above, in our study we also found some statistically significant associations between serum 25(OH)D quartiles and cardiovascular and AIS risk factors [11–13, 25, 35, 41], such as diabetes mellitus, blood glucose, HbA1c, atrial fibrillation, and hypertension (Table 1, Table 2). The relationships between serum 25(OH)D quartiles and plasma lipid concentrations were statistically significant but appear to be of limited clinical importance (Table 2). Such unfavorable associations between low serum vitamin D concentration and the prevalence of diabetes mellitus [25, 41], hypertension [35], and atrial fibrillation [35, 36–39] have been reported previously. They can be explained by previously reported associations between low vitamin D levels and obesity and related metabolic disorders, as well as increased insulin resistance, endothelial dysfunction, oxidative stress, chronic inflammation (elevated hs-CRP, CRP, IL-6, and TNF-α), and dysregulation of the renin–angiotensin–aldosterone system [35].

Reviewer 2 Report

Comments and Suggestions for Authors

Switonska et al. submit a retrospective study on stroke victims in a Polish hospital, with data collected between 2020 and 2025. 1,381 patients were classified according to their blood levels of 25(OH)-vitamin D and correlated to stroke mortality. In addition, numerous other parameters were used for correlational studies including parameters related to heart function, metabolism, blood cells and plasma lipids. Much of the results confirm previous studies, i.e. low vitamin D is associated with a poorer outcome of stroke. Some data hint to an effect of vitamin D dampening inflammatory responses. However, the interpretation of the data leaves some doubts as detailed below.

A major problem is that the authors list low p-trend values when there is no obvious trend in the data. In the first two lines of Table 1, for instance, the quartiles list as age: 70, 68, 67 and 70 years. How can the p-trend be <0.001? In the second line, male gender is 37, 44, 47 and 36%. There is no obvious trend but again p<0.001. Similar problems can be found with many other parameters, for instance atrial fibrillation, ejection fraction or readmission within 14 days. Please double-check your statistics. In Table 2, p-trends do not make sense with hemoglobin or LDL-cholesterol.

What I can see from the data is that Q1, the quartile of patients with low 25-OH-D levels (<10 ng/ml), has several poor health parameters. In this patient group, diabetes is more frequent and hyperglycemia and HbA1c levels are higher, hypertension and chronic kidney disease are more prevalent, as is (not surprisingly) all-cause in-hospital death. If patients survive, the NIHSS score upon discharge is higher than in patients with more vitamin D. Table 2 shows that Q1 patients have more neutrophils. Interestingly, total cholesterol is low. There seems to be a trend towards lower neutrophiles with higher vitamin D which support the idea of anti-inflammatory actions. I am not so impressed with the various ratios because they often overestimate changes based on basic data that have a wide margin of error. All in all, I cannot see much difference between Q2, Q3 and Q4.

Taken together, the manuscript is of interest but needs some clarification. The discussion is reasonable. Vitamin D supplementations make sense when levels are low. Vitamin D supplementation may not prevent stroke, but it seems to improve outcomes. I am not sure, however, if calcium plays a role in this respect, i.e. for effects in immune cells.

Minor points:

The article is full of abbreviations, not all of them are explained in the text. A list of abbreviations should be added, particularly for non-clinicians.

Line 283, mortality cannot be expressed in ng/ml.

The list of references is a bit long. It is not necessary to list ten individual publications if a review is cited. 

Author Response

1. A major problem is that the authors list low p-trend values when there is no obvious trend in the data. In the first two lines of Table 1, for instance, the quartiles list as age: 70, 68, 67 and 70 years. How can the p-trend be <0.001? In the second line, male gender is 37, 44, 47 and 36%. There is no obvious trend but again p<0.001. Similar problems can be found with many other parameters, for instance atrial fibrillation, ejection fraction or readmission within 14 days. Please double-check your statistics. In Table 2, p-trends do not make sense with hemoglobin or LDL-cholesterol.

Thank You for comment. We checked our statistics, and we did not find any mistakes. We corrected heading in controversial column in Table 1 and Table 2 into “p F test, Chi² test”. It seems to be clear now that this column contain p- value not for trend, which sometimes seems to more U-, J- or L- shaped than linear, but only for statistical significance of ANOVA F test (continuous variables) Chi² (categorical variables). We corrected respective comments to the Table 1 and Table 2 in the text of manuscript and abstract. 

 

2. What I can see from the data is that Q1, the quartile of patients with low 25-OH-D levels (<10 ng/ml), has several poor health parameters. In this patient group, diabetes is more frequent and hyperglycemia and HbA1c levels are higher, hypertension and chronic kidney disease are more prevalent, as is (not surprisingly) all-cause in-hospital death. If patients survive, the NIHSS score upon discharge is higher than in patients with more vitamin D. Table 2 shows that Q1 patients have more neutrophils. Interestingly, total cholesterol is low. There seems to be a trend towards lower neutrophiles with higher vitamin D which support the idea of anti-inflammatory actions. I am not so impressed with the various ratios because they often overestimate changes based on basic data that have a wide margin of error. All in all, I cannot see much difference between Q2, Q3 and Q4.

In Table 2- white blood cells (9.7 à 9.2 G/l) and neutrophils (7.7à 6.27 G/l) counts decreased with increase of vitamin D quartiles. We agree that such differences may to be not clinically relevant, however they are statistically significant, especially in regard to neutrophils count.

In discussion section we present some comment concerning the relationships between vitamin D quartile and indices of inflammatory response to brain injury. We checked also correlations between these biomarkers. For example:

Moreover, to the best of our knowledge, our study is the first to identify associations between serum 25(OH)D concentration and values of known and validated indices of inflammatory response to brain injury [3], such as the CRP, neutrophil-to-lymphocyte ratio, CRP-to-platelet ratio, SII, SIRI, and NPS (Table 2). This observation might suggest the real-world importance of 25(OH)D activity as a modulator of systemic, and probably cerebral, inflammatory response to AIS [3, 4, 35]. However, the power of the associations observed was weak and with low determination coefficient (R²) values. The relationships showed that serum 25(OH)D concentration determines only 1.7-3.6% of variances in the indices mentioned. This observation only weakly corroborates the results of a study in rats showing vitamin D as a strong immunomodulator mitigating the influence of brain ischemia/reperfusion-induced systemic and local inflammation [35, 45]. Nevertheless, we confirmed, known from literature [27], the importance of serum 25(OH)D concentration as an independent factor increasing the probability of a low disability score (mRS ≤ 2) at discharge on average by 52% for every quartile (Q1 à Q4), along with the neutrophil-to-lymphocyte ratio, a known and validated index of systemic inflammatory response to brain injury [3, 4], older age, diabetes mellitus, and treatment with EMT as a source of ischemia- reperfusion-induced neuroinflammation.

3. Vitamin D supplementations make sense when levels are low. Vitamin D supplementation may not prevent stroke, but it seems to improve outcomes. I am not sure, however, if calcium plays a role in this respect, i.e. for effects in immune cells.

Thank you for comment. We agree, in accordance with literature data, that vitamin D supplementation in patients with acute ischemic stroke is reasonable in patients with low vitamin level. We present also different cut-offs values for vitamin target level to improve patient’s survival and functional impairment, as below. Calcium supplementation was mentioned in Discussion section only as potential cause of adverse effect of vitamin D supplementation, which is hypercalcemia.

Interestingly, the ROC analysis indicated that the 25(OH)D level predicting all-cause in-hospital mortality in AIS patients was lower than that determined for a low disability score (mRS ≤ 2) at discharge (< 20.48 vs. > 46.86 ng/ml). This may be important for the dosing and monitoring of vitamin D supplementation according to its intended purpose, as well as for the design of future studies and the planning of outcome measures.

 

4. The article is full of abbreviations, not all of them are explained in the text. A list of abbreviations should be added, particularly for non-clinicians.

 

In our opinion all abbreviations were explained in the text. We added, abbreviation list at the end of manuscript.

 

5. Line 283, mortality cannot be expressed in ng/ml.

We corrected this.

6. The list of references is a bit long. It is not necessary to list ten individual publications if a review is cited. 

The number of references cited was reduced. Citation correctness was checked.

Reviewer 3 Report

Comments and Suggestions for Authors

Major Comments:
The conclusion that vitamin D has neuroprotective properties or functions as a "immunomodulator" may be exaggerated because the study is observational and retrospective. Throughout the text, the manuscript should prioritize association over causality. Multicenter prospective validation is required.

A number of significant confounders, including seasonality, sun exposure, nutritional status, BMI, smoking, socioeconomic status, physical activity, and chronic inflammatory conditions—all of which have a significant impact on serum 25(OH)D levels and stroke outcomes—were not adequately controlled despite the use of multivariate logistic regression.

The report does not specify how supplementing history was confirmed, however patients who were taking vitamin D supplements were eliminated. There is no information about the duration, dosage, or adherence to previous over-the-counter supplements.

Patients are not categorized in the study based on their ischemic stroke subtype (cardioembolic, big artery atherosclerosis, lacunar, etc.) or TOAST classification. Vitamin D relationships may be impacted by the different inflammatory profiles and prognosis outcomes associated with different stroke etiologies.

The precise time of serum 25(OH)D assessment in relation to stroke onset is not precisely disclosed, despite the manuscript stating that laboratory data were obtained throughout hospitalization. Vitamin D and inflammatory biomarker levels may change due to acute sickness.

Although only indirect hematologic ratios were examined, the study interprets inflammatory markers as proof of vitamin D's immunomodulatory effect. The suggested immunological mechanism was not supported by any cytokines, immune cell phenotyping, oxidative stress markers, or mechanistic experiments.

Numerous inflammatory indicators and quartile analyses were subjected to multiple comparisons without multiple testing or false discovery rate correction. This raises the risk of a type I error.

The published AUC values, such as AUC 0.587 and 0.545, are comparatively low, suggesting that serum vitamin D content is not a very good clinical biomarker. The limited clinical value of these findings should be discussed more seriously.

In addition to being younger, patients with higher vitamin D levels also had less comorbid conditions like diabetes and chronic kidney disease. Despite regression analysis, these baseline variations could have a substantial impact on mRS results.

Minor Comments:
The title should be revised to:
“Serum 25-Hydroxyvitamin D Concentration as a Biomarker and Immunomodulator in Patients with Acute Ischemic Stroke: A Retrospective Single-Center Study.”

The document makes extensive use of a number of acronyms. Readability would be enhanced with a distinct list of abbreviations or more lucid first-use definitions.

There are a few punctuation and spacing errors, especially in the areas of references, hyphenation, and statistical notation.

Excessive statistical comparisons make Tables 1 and 2 difficult to understand. Think about transferring certain pairwise comparisons to supplemental resources or making tables simpler.

The Discussion section has a number of extremely long and confusing sentences. Clarity and readability would be enhanced by expert English editing.

There isn't a specific restrictions section in the manuscript. Retrospective design, lack of long-term vitamin D assessments, lack of dietary evaluation, lack of long-term neurological follow-up, and residual confounding are significant limitations that should be clearly mentioned.

Author Response

1. The conclusion that vitamin D has neuroprotective properties or functions as a "immunomodulator" may be exaggerated because the study is observational and retrospective. Throughout the text, the manuscript should prioritize association over causality. Multicenter prospective validation is required.

 

We reduced the strength of conclusion prioritizing association over causality, for example:

Higher serum vitamin D concentration in AIS patients was associated with lower all-cause mortality, a lower level of disability at discharge from hospital, and a lower risk of readmission within one year (equivalent to a lower risk of stroke recurrence). The association may be explained by vitamin D activity controlling the severity of cardiovascular risk factors and mitigating the inflammatory response to brain injury.

 

2. A number of significant confounders, including seasonality, sun exposure, nutritional status, BMI, smoking, socioeconomic status, physical activity, and chronic inflammatory conditions—all of which have a significant impact on serum 25(OH)D levels and stroke outcomes—were not adequately controlled despite the use of multivariate logistic regression.

We added information concerning clinical relevance of confounders mentioned and necessity to careful interpretation of the results obtained in study limitation section, as below:

It is important to recognize that our study has a few limitations that need to be identified and addressed in future investigations. First, this study was conducted in a single center in Poland and the lack of an external validation study cohort may limit the generalizability of our findings to other ethnic and geographic populations, associated with potential confounders, including sun exposure, seasonality, socioeconomic, general health and nutritional status, liver and kidney function, pre-stroke physical activity, vitamin D supplementation, comorbidities, changes in vitamin D level due to acute sickness, delay in blood sampling due to endovascular treatment, etc. [35].

3. The report does not specify how supplementing history was confirmed, however patients who were taking vitamin D supplements were eliminated. There is no information about the duration, dosage, or adherence to previous over-the-counter supplements.

Thank You for comment. We agree that unrecognized vitamin D supplementation may affect the results obtained. We took it into consideration during analysis of available data in patient medical documentation, focusing on documentation of pre-stroke medication. Unfortunately, due to retrospective study design we cannot avoid some shortcomings. Therefore, we completed the text:

 

The following exclusion criteria were applied: known supplementation of vitamin D (in medical history obtained and in patient medication list), ….

Moreover, this study was designed as a retrospective analysis of medical documentation, which should be recognized as an important study limitation even though it provides real-world data. One important potential limitation of such an analysis is e.g., recall bias regarding vitamin D supplementation before admission, although we analyzed pre-stroke medical treatment and supplementation on the basis of both the medical interview and the patient's medication list.

 

4. Patients are not categorized in the study based on their ischemic stroke subtype (cardioembolic, big artery atherosclerosis, lacunar, etc.) or TOAST classification. Vitamin D relationships may be impacted by the different inflammatory profiles and prognosis outcomes associated with different stroke etiologies.

The analysis concerning patient TOAST categorization was added to the text of manuscript, and in Table 1.

In Patients and Methods section: The presumed etiological mechanism of AIS was classified according to the Trial of Org 10172 in Acute Stroke Treatment (TOAST) classification, distinguishing the following stroke subtypes: large-artery atherosclerosis (LAA) plaque buildup, cardioembolic stroke (CE), small-vessel occlusion (SVO)/lacunar stroke, and stroke of undetermined etiology (SUE).

In the Results section: With regard to the TOAST classification of presumed AIS etiology, the study included 432 patients (31.3%) with LAA, 443 (32.0%) with CE, 172 (12.5%) with SVO, and 334 (24.2%) with SUE. Serum 25(OH)D concentration did not differ significantly among the TOAST categories (22.0 ± 15.8 vs. 20.8 ± 15.3 vs. 21.5 ± 14.4 vs. 21.5 ± 16.0 ng/ml, respectively; ANOVA, F = 0.433; p = 0.729). Moreover, although the overall chi-square (χ²) test showed a statistically significant association between TOAST category and 25(OH)D quartile, only a single statistically significant difference was found across the etiological stroke categories and 25(OH)D quartiles (Table 1).

In Discussion section: In our study we did not confirm the relationships between serum 25(OH)D level and TOAST etiological stroke categorization, however, we found also no publication which showed that low vitamin level is associated with single AIS subtype.

 

5. The precise time of serum 25(OH)D assessment in relation to stroke onset is not precisely disclosed, despite the manuscript stating that laboratory data were obtained throughout hospitalization. Vitamin D and inflammatory biomarker levels may change due to acute sickness.

Blood sampling for biochemical determination was taken after patient admission to the clinic, however serum 25(OH)D assessment in this sample was performed morning next day. We added this study shortcoming to the study limitation list.

…. vitamin D supplementation, comorbidities, changes in vitamin D level due to acute sickness, differences in time of blood sampling in regard to stroke onset due to delay in patient admission, transport, and endovascular treatment, etc. [35].

 

6. Although only indirect hematologic ratios were examined, the study interprets inflammatory markers as proof of vitamin D's immunomodulatory effect. The suggested immunological mechanism was not supported by any cytokines, immune cell phenotyping, oxidative stress markers, or mechanistic experiments.

Thank You for comment. We enumerated this as study limitation:

Second, we only collected baseline levels of 25(OH)D and indices of inflammatory response to brain injury on admission, which were also not validated by determination of any cytokines, immune cell phenotyping, oxidative stress markers.

7. Numerous inflammatory indicators and quartile analyses were subjected to multiple comparisons without multiple testing or false discovery rate correction. This raises the risk of a type I error.

To avoid type I error related to multiple comparisons we used ANOVA analysis with post-hoc test.

 

8. The published AUC values, such as AUC 0.587 and 0.545, are comparatively low, suggesting that serum vitamin D content is not a very good clinical biomarker. The limited clinical value of these findings should be discussed more seriously. In addition to being younger, patients with higher vitamin D levels also had less comorbid conditions like diabetes and chronic kidney disease. Despite regression analysis, these baseline variations could have a substantial impact on mRS results.

 

Thank You for comment, however we raised poor predictive power of our ROC analysis.

 

RESULTS section [line 297-304]:

In ROC analysis, we found only poor predictive power between serum 25(OH)D concentration and all-cause in-hospital mortality (cut-off: 20.48 ng/ml; area under the curve [AUC], 95% confidence interval [CI]: 0.587, 0.532-0.642; p = 0.002); readmission within 365 days after discharge (cut-off: 19.71 ng/ml; AUC, 95% CI: 0.545, 0.514-0.577; p = 0.005); and low disability score at discharge, of mRS ≤ 2 (cut-off: 46.86 ng/ml; AUC, 95% CI: 0.409, 0.357-0.467; p = 0.002). However, a serum 25(OH)D concentration equal to or greater than 46.86 ng/ml concerned only 97 (7%) of the patients studied.

 

Discussion section [line 457-467]

Our study also identified target serum 25(OH)D concentrations associated with patient survival (> 20.5 ng/ml) and with low disability at discharge (> 47 ng/ml); the concentration associated with low disability exceeds the target level reported for cardiovascular event prevention [41]. On the other hand, it should be underlined that AUC obtained in ROC analysis was low (0.587 for all-cause mortality, and 0.509 for mRS ≤ 2), what may suggest also that determination of serum 25(OH)D concentration may not be a good biomarker of poor and favorable AIS treatment outcomes. Such statement is supported by the fact that serum vitamin D concentrations above cut-offs determined were associated with younger age and lower comorbidity load among patients with better prognosis (Table 1).

 

9. 
   The title should be revised to: “Serum 25-Hydroxyvitamin D Concentration as a Biomarker and Immunomodulator in Patients with Acute Ischemic Stroke: A Retrospective Single-Center Study.”

It was done.

10. The document makes extensive use of a number of acronyms. Readability would be enhanced with a distinct list of abbreviations or more lucid first-use definitions.

All abbreviations used in the text of manuscript are explained during their firs use (highlighted in red). Moreover, the list of abbreviations was added to the text

Abbreviation list:

25(OH)D = vitamin D

CALLY = CRP-albumin-lymphocyte

CRP = C- reactive protein

HALP = hemoglobin-albumin-lymphocyte-platelet

HbA1c = hemoglobin A1c

HLAN = lymphocyte-albumin-neutrophil

IBI = Inflammatory Burden Index (CRP, NLR)

LMR = lymphocyte-to-monocyte ratio

mRS = modified Rankin Scale;

mTICI = modified Thrombolysis In Cerebral Infarction scale;

NIHSS = National Institute of Health Stroke Scale.

NLR = neutrophil-to-lymphocyte ratio

NPAR= neutrophil percentage-to-albumin ratio

NRI = The Nutritional Risk Index

PHR = platelet-to-hemoglobin ratio

SII = Systemic Immune-Inflammation Index

SIRI = Systemic Inflammation Response Index

TOAST = Trial of Org 10172 in Acute Stroke Treatment: CE = cardioembolic stroke, LAA = large artery atherosclerosis; SVO = small vessel occlusion; SUE = stroke of undetermined etiology

 

 

11. There are a few punctuation and spacing errors, especially in the areas of references, hyphenation, and statistical notation.

We corrected what was possible. Problems with hyphenation are independent of the authors, and was imposed by Editorial Office.

 

12. Excessive statistical comparisons make Tables 1 and 2 difficult to understand. Think about transferring certain pairwise comparisons to supplemental resources or making tables simpler.

Thank you for comment. We did not change the tables because only one Reviewer raised this issue. We think that replacement of p-value with graphic signs, for example: “*”, “+”, “#”might make Tables simpler, but may make more difficult to find significant differences, when significance of differences would be multiple, e.g. “*++#”

 

13. The Discussion section has a number of extremely long and confusing sentences. Clarity and readability would be enhanced by expert English editing.

The manuscript was checked by native speaker

 

14.There isn't a specific restrictions section in the manuscript. Retrospective design, lack of long-term vitamin D assessments, lack of dietary evaluation, lack of long-term neurological follow-up, and residual confounding are significant limitations that should be clearly mentioned.

We added this to the text:

The lack of serial (dynamic) measurements and neurological follow-up made it impossible to monitor changes in inflammatory indices over time and identify which is the strongest in relation to admission serum vitamin D concentration, in order to determine the role of serum vitamin D concentration as a systemic modulator and, therefore, perhaps as a modulator 

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

The authors have provided comprehensive responses to all reviewer comments and have made the necessary revisions throughout the manuscript.

Author Response

Thank You very much

Reviewer 2 Report

Comments and Suggestions for Authors

Several comments have been addressed but the most important point, my doubt about the statistics, has not been addressed, and the p values were not changed. I do not believe the statistics in the tables. One example (Table 1, first line): when you compare 70.66 +/- 12.68 with 67.09 +/- 13.56 years, the p value cannot be P=0.000. One does not need a calculator to see this.  

Author Response

D

Please find attached original table of statistics (sorry, but originally in polish), Student t-test and ANOVA analysis with post-hoc test. I cannot prove in another way that our statistics are reliable and true.

Zmienna	Testy t; Grupująca: QD3: kwartyle D3, 10;16, 17,78, 29.01 (i63_2025-09-19 w
	Średnia 1	Średnia 3	t	df	p	Nważnych 1	Nważnych 3	Odch.std 1	Odch.std 3	iloraz F Wariancje	p Wariancje
age	70,66279	67,08934	3,576812	689	0,000372	344	347	12,67868	13,56447	1,144610	0,210654

 

ANOVA analysis

Nr podkl.	"QD3"; Oczekiwane średnie brzegowe (i63_2025-09-19 w NUTRISTROKE_OST) Bieżący efekt: F(3, 1377)=4,9179, p=,00211 Dekompozycja efektywnych hipotez
	QD3	age Średnie	age Bł. Std.	age -95,00%	age +95,00%	N
1	1	70,66279	0,708014	69,27389	72,05169	344
2	2	68,97093	0,708014	67,58203	70,35983	344
3	3	67,08934	0,704947	65,70645	68,47222	347
4	4	70,04046	0,705965	68,65558	71,42535	346

 

And post- hoc test

Nr podkl.	Test NIR; zmienna wiek (i63_2025-09-19 w NUTRISTROKE_OST) Prawdopodobieństwa dla testów post-hoc Błąd: MS międzygrupowe = 172,44, df = 1377,0- AGE
	QD3	AGE 70,663	{AGE 68,971	AGE 67,089	AGE 70,040
1	1		0,091313	0,000360	0,533762
2	2	0,091313		0,059876	0,284938
3	3	0,000360	0,059876		0,003149
4	4	0,533762	0,284938	0,003149

 

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

The author improve significantly.

Author Response

Thank You very much