Association between hematocrit and the 30-day mortality of patients with sepsis: A retrospective analysis based on the large-scale clinical database MIMIC-IV.

This research sought to ascertain the relationship between hematocrit (HCT) and mortality in patients with sepsis.
METHODS: A retrospective analysis was conducted on the clinical data of septic patients who were hospitalized between 2008 and 2019 in an advanced academic medical center in Boston, Massachusetts, registered in the Medical Information Mart for Intensive Care IV (MIMIC-IV) database, We analyzed basic information including gender, age, race, and types of the first admission, laboratory indicators including HCT, platelets, white blood cells, albumin, bilirubin, hemoglobin, and serum creatinine, and 30-day mortality. A Cox proportional hazards regression model was utilized to analyze the relationship between HCT and 30-day mortality in patients with sepsis.
RESULTS: This research recruited 2057 patients who met the research requirements from 2008 to 2019. According to the HCT level, it was classified into the low HCT level, the regular HCT level, and the high HCT level. The 30-day mortality rate was 62.6%, 27.5%, and 9.9% for patients with the low HCT level, the regular HCT level, and the high HCT level, respectively (p < 0.05). The multivariate Cox proportional hazard regression model analysis displayed that compared with patients with the regular HCT level, the 30-day mortality of patients with the low HCT level increased by 58.9% (hazard ratio = 1.589, 95% confidence interval (CI) = 1.009-2.979, p < 0.05).
CONCLUSION: The low HCT level is an independent risk factor for the increase of the 30-day mortality in patients with sepsis and can be used as a significant predictor of the clinical outcome of sepsis.

Introduction

Sepsis is a lethal syndrome of physiologic, pathologic, and biochemical abnormalities induced by infection, which is one of the major global public health concerns [1]. Although recently there has been extensive researches demonstrating the mechanism and treatment of sepsis, sepsis is still the principle cause of death in intensive care patients worldwide [2]. There are a variety of scores for the diagnosis of sepsis that can be evaluated, but there is still a lack of valuable indicators for the study on prognostic factors [3]. Anemia is one of the risk factors for death resulting from sepsis and septic shock, and hematocrit (HCT) is the percentage of red blood cells in the volume of the whole blood, which is one of the critical biomarkers for the diagnosis of anemia [4].

HCT can be utilized as a critical prognostic biomarker in several cancers, including lung cancer, renal carcinoma, and epithelial ovarian cancer. However, little is known about the prognostic value of HCT for patients with sepsis in surgical intensive care units (ICUs) [5]. At present, clinical researches have mainly focused on the relationship of anemia indicators such as the red blood cell distribution width (RDW) and platelets with the prognosis of sepsis [6]. In spite of limited researches on the impacts of HCT on the prognosis of patients with sepsis, a few studies are limited to the evaluation of anemia before sepsis surgery [7]. Accordingly, the HCT level of first admission patients was employed in our research to dissect out whether it influenced the prognosis of sepsis, thus helping doctors assess the condition in a timely manner and provide a basis for subsequent prognostic measures.

Method

Data source

All data used in the study was extracted from the MIMIC-IV (v1.0) database, which is the hospitalization information of patients admitted to the Higher Medical Center in Boston, Massachusetts, the USA from 2008 to 2019 to conduct a retrospective study. This database is a relational database containing the actual hospitalizations of hospitalized patients in an advanced academic medical center in Boston, Massachusetts, USA. The MIMIC-IV database is based on the success of MIMIC-III [8]. It integrates improvements to the deficiencies of the MIMIC-III database, including laboratory measurements, medications, recorded vital signs, SOFA score, SAPS II score, etc.

Study population and data extraction

PgAdmin4 has used run structure query language (SQL) and extracted data from the MIMIC IV database. Patients are identified by ICD-9 codes and extracted from the database [9]. The extracted variables included age, gender, admission type, comorbidities, laboratory parameters, severity score, time of entering and leaving the hospital, time of entering and leaving the ICU, and date of death. Laboratory parameters include hematocrit, albumin, lactate, bilirubin, hemoglobin, red blood cell distribution width, platelet, serum creatinine, potassium, etc. All laboratory data were extracted from the data generated within the first 24 hours after the patient entered the ICU (i.e., the baseline value). Data in this retrospective analysis is accurate medical data, accessed free of charge. The project has been approved by the Institutional Review Board of the Massachusetts Institute of Technology and BIDMC and has received an informed consent exemption. The author has the right to use and download the database through the Protecting Human Research Participants exam.

Patient population

Adult patients (age ≥18 years) admitted to ICU (internal medicine ICU, surgical ICU) who had the exact hematocrit index and corresponding ID within 24 hours after admission to the ICU. Patients who lack corresponding data will be excluded. Sepsis criteria: This study uses patients diagnosed with sepsis in the database and meets sepsis-3 to define the sepsis cohort.

Statistical analysis

Analyses were performed using Stata 16. Baseline characteristics of all patients were stratified according to the quartiles of hematocrit (HCT) and divided into three groups: low hct level (male≤42%, Female≤37%), regular hct level (male 42%~49%, female 37%~44%), and high hct level (male≥49%, female≥44%) [4]. Continuous variables were presented as median, and categorical variables were presented as a number. Nonparametric Wilcoxon tests or Kruskal-Wallis tests were used to compare continuous variables between different groups. Categorical variables between the groups were compared using the X2 test and Fisher exact test. A P value < 0.05 between the two groups was considered a significant difference. A Multivariate Cox proportional hazard model was constructed to determine the independent effects of three groups on 30-day mortality. Variables with P <0.05 in univariate analysis were further included in the multivariate Cox proportional hazard model. We compared the survival rates using log-rank tests and presented the results as Kaplan-Meier curves. A Kaplan-Meier survival curve was constructed to compare the 30-day mortality of the three groups of hct levels. P<0.05 indicates that the difference is statistically significant.

Result

Population and baseline characteristics

A total of 2057 patients with sepsis 3.0 in the MIMICIV database were included which have complete data of hematocrit level within 24 hours after ICU admission. All baseline characteristics are summarized in Table 1. Differences in age, laboratory parameters, comorbidities, and scores between the two groups were statistically significant. HCT level was divided into three groups. Variables with missing data are relatively common in the MIMIC IV database. The number of deaths within 30 days after admission to the ICU was 911. Compared with the 30-day survival group, the death group was older, and the levels of hematocrit, albumin, platelets, and hemoglobin were lower than those in the survival group (P<0.05).

Table 1

Characteristics of the patients with sepsis.

Variables	30-day mortality, n = 911 1	30-day survival, n = 1146 0	P
Age, years	70[59,81]	66[54,79]	<0.001
Female, n (%)	418 (45.88)	519 (45.29)	0.788
First care unit, n (%)			0.035
MICU	294 (32.34)	398(34.88)
MICU/SICU	254 (27.94)	342 (29.97)
SICU	216 (23.76)	263 (23.05)
Others	145 (15,95)	138 (12.09)
Mechanical Ventilation, n (%)	247[27.11]	232[20.24]	<0.001
Comorbidities, n (%)
Hypertension, n (%)	106 (11.64)	193 (16.94)	<0.001
COPD, n (%)	91 (9.99)	93 (8.12)	0.139
Diabetes, n (%)	46 (5.05)	59 (5.15)	<0.001
Respiratory failure, n (%)	374 (41.05)	309 (26.96)	0.014
CHF, n (%)	42 (4.61)	30 (2.62)	<0.001
Vital signs
SBP, mmHg	113.41[105.0,119.2]	116.3[110.3,119.5]	<0.001
DBP, mmHg	57.0[51.0,60.3]	58.4[55.8,61.7]	<0.001
Heart rate, bmp	94.0[76.0,120.0]	94.0[71.0,120.0]	0.093
MAP, mmHg	72.0[63.0,79.5]	76.8[70.0,80.0]	<0.001
Laboratory parameters
Albumin, g/dl	3.0[2.5,3.6]	3.2[2.7,3.7]	<0.001
White blood cell,109/L	14.2[10.8,17.9]	14.2[11.3,18.0]	0.507
Hematocrit,%	35.0[30.0.39.9]	35,9[31.0,40.8]	<0.001
Platelet,109/L	153.0[91.0.230.0]	181.0[125.0,255.0]	<0.001
Bilirubin, mg/dL	1.3[0.6,3.1]	0.9[0.4,2.8]	<0.001
Hemoglobin, g/dL	9.6[9.4,9.9]	9.6[9.5,9.8]	0.037
RDW,%	14.6[13.6,16.2]	13.9[13.0,15.1]	<0.001
Scr, mg/dl	1.6[1.0,2.6]	1.3[0.6,2.1]	<0.001
Lactate, mmol/L	2.5[1.6,4.6]	1.9[1.3,2.8]	<0.001
Sodium, mEq/L	138.0[134.0,142.0]	139[141.0,135.0]	0.186
Potassium, mEq/L	4.2[3.7,4.8]	4.0[3.6,4.5]	<0.001
Severity scores
SOFA	6.0[3.0,9.0]	4.0[3.0,6.0]	<0.001
SAPSII	52.0[42.0,63.0]	42.0[33.0,52.0]	<0.001

*Data are expressed as median (IQR), or n (%). Analysis of variance (or the Kruskal-Wallis test) and Chi-square (or Fisher’s exact) tests were used for comparisons among groups. Statistical significance (P<0.05).

ICU, intensive care unit; SICU, Surgical Intensive Care Unit; MICU, Medicine Intensive Care Unit; COPD, Chronic obstructive pulmonary disease; CHF, Congestive heart failure; SBP, Systolic blood pressure; DBP, Diastolic blood pressure; MAP, mean arterial pressure; RDW, Red blood cell distribution width; Scr, Serum creatinine; SOFA, Sequential Organ Failure Assessment; SAPSII, Simplified Acute Physiology Scores II.

Association of Hematocrit level with 30-day outcomes

The 30-day all-cause mortality rate of patients was 44.2%, As shown in Table 2, compared with the survival group, the proportion of patients with low hct levels in the 30-day mortality group was significantly increased, and the proportion of the regular group and the high hct level group gradually decreased. The difference was statistically significant (P<0.05).

Table 2

Outcomes of sepsis patients according to the hematocrit level.

	Low hematocrit level	Regular hematocrit level	High hematocrit level	P
30-day mortality, n (%)	569(62.63)	251(27.55)	91(9.99)	<0.05
30-day survival, n (%)	653(56.98)	358(31.24)	135(11.78)	<0.05
Total (n = 2057), n (%)	1222(59.41)	609(29.61)	226(10.99	<0.05

Hematocrit is an independent prognostic predictor in sepsis patients

Survival analysis was conducted to explore the impact of hct level on 30d mortality. Notably, from the previous analysis, we know that patients in the lower hct level had worse survival rates. Basic demographics and laboratory parameters for the prediction of 30-day mortality were investigated using a univariate Cox analysis regression model. Variables include age, low hct levels, albumin, platelets, bilirubin, red blood cell distribution width, lactate, Scr, SBP, DBP MAP, SAPS II score, hypertension, heart failure, and mechanical ventilation are all statistically significant (p<0.05). Adjust the univariate analysis for the potential confounding factors associated with 30-day mortality in patients with sepsis, And then, the 30-day mortality was assessed with a multivariable Cox proportional regression model. According to the results, the low hct level remained an independent prognostic factor for sepsis (P<0,05). Compared with the regular hct level, the 30-day mortality risk of the low hct level is increased, and the difference is statistically significant [HR = 1.589 .95%CI1.099–2.297, P<0.05], although the risk of death in the high hct level also increased, the difference was not statistically significant (P = 0.055) (Table 3).

Table 3

Univariate and multivariable analyses for the relationship between the candidate risk factors and 30- day mortality in the primary cohort.

Variables	Univariate model			Multivariable model
	HR	95%CI	P	HR	95%CI	P
Age	1.018	1.014    1.023	<0.001	1.016	1.011    1.022	<0.001
Female	1.105	0.970    1.259	0.133
MAP	0.996	0.994    0.998	<0.001	0.998	0.995    0.999	0.025
DBP	0.987	0.981    0.992	<0.001	0.995	0.996    1.000	0.038
SBP	0.995	0.982    0.992	<0.001	0.994	0.988    1.000	0.071
Rerular hct level	Reference <0.001
Low hct level	1.369	1.116    1.745	0.003	1.589	1.099    2.297	<0.001
High hct level	1.155	0.908    1.470	0.243	1.330	1.007    1.757	0.055
RDW	1.120	1.091    1.149	<0.001	1.072	1.040    1.105	<0.001
Potassium	1.078	0.998    1.165	0.057
Scr	1.066	1.027    1.105	0.001	1.014	0.971    1.059	0.519
Albumin	0.915	0.841    0.994	0.037	0.941	0.859    1.031	0.191
Hemoglobin	0.986	0.929    1.009	0.121
Platelet	1.000	0.998    1.000	<0.001	0.999	0.999    1.000	0.034
Bilirubin	1.035	1.026    1.045	<0.001	1.026	1.014    1.038	<0.00
Lactate	1.123	1.099    1.148	<0.001	1.087	1.061    1.114	<0.001
SOFA	1.002	0.985    1.020	0.789
SAPSII	1.020	1.016    1.024	<0.001	1.011	1.006    1.016	<0.001
Diabetes	1.014	0.754    1.365	0.952
Hypertension	0.788	0.643    0.965	0.021	0.627	0.507    0.775	<0.001
COPD	1.109	0.958    1.478	0.116
Respiratory failure	0.950	0.831    1.086	0.455
CHF	1.666	1.221    2.273	0.001	1.357	0.986    1.866	0.061
Mechanical ventilation	1.981	1.703    2.305	<0.001	0.460	0.393    0.540	<0.001

hct, hematocrit; COPD, Chronic obstructive pulmonary disease; CHF, Congestive heart failure; SBP, Systolic blood pressure; DBP, Diastolic blood pressure; MAP, mean arterial pressure; RDW, Red blood cell distribution width; Scr, Serum creatinine; SOFA, Sequential Organ Failure Assessment; SAPSII, Simplified Acute Physiology Scores II.

Kaplan-Meier survival curve analysis

The Kaplan-Meier survival curve was drawn according to the category of hct to show the 30-day survival rate of patients with sepsis. The results showed that the difference between HCT level and the 30-day mortality rate of sepsis was statistically significant (P = 0.040), red blood cell ratio Content is related to the prognosis of patients with sepsis (Fig 1).

Fig 1

Survival curves showing the association between the hct level and 30-day mortality.

Discussion

Recently, sepsis remains the main cause of death triggered by the infection in ICUs despite of considerable breakthroughs in the exploration of sepsis, including broad-spectrum antibiotics, supportive treatment, and even precision medicine and monitoring [10]. Infection and immune response disorders have been recognized as risk factors for organ dysfunctions. The poor functional status has been reported as a risk factor for sepsis and an expected consequence based on the long-term investigation of sepsis over the past decade [11, 12]. Therefore, it is essential to be able to early predict and evaluate the prognostic factors for sepsis. Infection is one of the crucial reasons for the occurrence and development of sepsis. This is because it can touch the systemic inflammatory response and cause the formation of inflammasomes in the body, thereby inducing the release of numerous pro-inflammatory factors [13]. Therefore, it is imperative to use effective biomarkers to monitor the prognosis of sepsis. Whole blood parameters have been partially studied as biomarkers in the diagnosis, treatment, and prognosis of sepsis. Whole blood viscosity, red blood cell aggregation, and red blood cell deformability may be risk factors for sepsis and septic shock mortality [14]. In addition, the study conducted by Gong Yan et al. [15] has elucidated that elevated RDW can remarkably predict disease progression and poor clinical outcomes in patients with sepsis. Currently, several emerging studies have been conducted to ascertain the influence of HCT on the prognosis of sepsis. It has been elaborated that HCT is correlated with all-cause mortality in end-stage renal disease, heart failure, coronary heart disease, cancer, and inflammatory states [16]. However, HCT has not been determined as an independent influencing factor for clinical outcomes of diseases. In this study, 2057 patients with sepsis were evaluated to figure out the correlation between the HCT level and the prognosis of sepsis. The results illustrated that compared with septic patients with the regular HCT level, the 30-day mortality rate was enhanced in septic patients with a low HCT level measured within 24 hours after entering the ICU (male ≤ 42% and female ≤ 37%), accompanied by conspicuously augmented Simplified Acute Physiology Score II scores. Following gradual regression and correction of various potential confounding factors, we obtained reliable results from the study, suggesting that HCT might be an independent risk factor for the prognosis of patients with sepsis, which is the same as the previous group of sepsis anemia patients and Western countries. The results of a small-scale clinical trial are concordant [7, 17].

HCT is a whole blood parameter that reflects the ratio between red blood cells and plasmas. It is closely linked to the prognosis of critically ill patients and indicators for the fluid resuscitation treatment. However, the relationship between HCT and the prognosis of patients with sepsis remains poorly understood. Acute systemic infections lead to inflammatory responses that result in sepsis, which dramatically diminishes the number of red blood cells entering the blood circulation [18]. The production of reactive oxygen species may contribute to the repression of the ability of red blood cells to transport oxygen and the deformities of red blood cell membranes. During the entire process of inflammatory response and oxidative stress, the number of red blood cells is diminished and the blood dilution induced by liquid expansion leads to a reduction in the HCT level [19-21]. It is necessary to further prove these research hypotheses. It has been documented that inflammatory factors like tumor necrosis factor α, interleukin (IL)-1, IL-2, IL-6, and IL-8 trigger the adhesion of neutrophils and endothelial cells, leading to the formation of microthrombi [22]. This mechanism may be related to the rapid removal of red blood cells from the circulating blood caused by inflammation and oxidative stress in sepsis, which is manifested by anemia and accelerated red blood cell apoptosis [23]. In addition, red blood cell infusion has been highlighted to improve oxygen transport and metabolism, thereby alleviating microcirculation disorders in patients with sepsis and septic shock [24]. Therefore, the hypotheses about the HCT were proposed on the basis of the aforementioned research. As reflected by the results of this study, patients with the low HCT level exhibit low platelets, hemoglobin, and the high 30-day mortality rate. The mechanism mentioned above can provide a basis for this theory. Paolo Boffetta et al. [16] unraveled that HCT is also associated with the mortality of ordinary people. A retrospective study by Zhang Xin et al. [25] unveiled that the low HCT also correlated to the poor prognosis of patients with lung cancer and ovarian cancer. These findings suggest that HCT may reflect the severity of a wide range of diseases.

This study was a large-scale retrospective study, in which plenty of factors related to the death of sepsis were harvested and adjusted to evaluate their impacts on the prognosis. The results manifested that HCT could be an independent risk factor for the prognosis of patients with sepsis. Although HCT is not highly specific according to our results, it is a simple laboratory parameter that is easier to obtain. The HCT levels obtained in our research were all information of patients with sepsis on the first day after entering the ICU, minimizing the changes in the HCT levels caused by the progression of the disease and the treatment process. However, there also exist several limitations in this study: (1) The Medical Information Market Intensive Care Database is a sizeable single-center database that lacks diversification. Therefore, our results are influenced by unity and may have inevitable bias. (2) Due to the singularity of the database, we can only conduct observational researches on HCT and mortality, so further studies are needed to understand the underlying pathophysiological mechanism in the future. (3) Taking the problem of the content of the data sample into account, the sample is not classified by gender, but a general study is carried out.

In summary, HCT is associated with the prognosis of septic patients during admission to the ICU. Septic patients with the low HCT level presented with enhanced disease severity and high mortality rate. HCT can be applied as an independent risk factor for the prognosis of patients with sepsis.

Further study is required to investigate the pathophysiological and immunological mechanism of the relationship between HCT and clinical outcomes in septic patients. A perfect biomarker for sepsis has not yet been identified. More immunological experiments and multi-center studies on this easily available parameter will be of great significance for the early prediction of the outcomes of sepsis, which must be beneficial for the treatment of sepsis.

Conclusion

In summary, the hematocrit in the first 24 hours after ICU admission was independently associated with increased 30-day all-cause mortality in adult septic patients but of limited sensibility and specificity. Further extensive multicenter prospective studies are needed to confirm the relationship and validate whose clinical significance.

14 Jan 2022

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Andrea Cortegiani, M.D.

Academic Editor

PLOS ONE

Journal Requirements:

1. When submitting your revision, we need you to address these additional requirements.

Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2.  Please include your tables as part of your main manuscript and remove the individual files. Please note that supplementary tables (should remain/ be uploaded) as separate "supporting information" files.

3. Thank you for stating the following financial disclosure:

(No)

At this time, please address the following queries:

a) Please clarify the sources of funding (financial or material support) for your study. List the grants or organizations that supported your study, including funding received from your institution.

b) State what role the funders took in the study. If the funders had no role in your study, please state: “The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.”

c) If any authors received a salary from any of your funders, please state which authors and which funders.

d) If you did not receive any funding for this study, please state: “The authors received no specific funding for this work.”

Please include your amended statements within your cover letter; we will change the online submission form on your behalf.

4. Thank you for stating the following in your Competing Interests section:

(No)

Please complete your Competing Interests on the online submission form to state any Competing Interests. If you have no competing interests, please state "The authors have declared that no competing interests exist.", as detailed online in our guide for authors at http://journals.plos.org/plosone/s/submit-now

This information should be included in your cover letter; we will change the online submission form on your behalf.

5.  We note that you have stated that you will provide repository information for your data at acceptance. Should your manuscript be accepted for publication, we will hold it until you provide the relevant accession numbers or DOIs necessary to access your data. If you wish to make changes to your Data Availability statement, please describe these changes in your cover letter and we will update your Data Availability statement to reflect the information you provide.

6. PLOS requires an ORCID iD for the corresponding author in Editorial Manager on papers submitted after December 6th, 2016. Please ensure that you have an ORCID iD and that it is validated in Editorial Manager. To do this, go to ‘Update my Information’ (in the upper left-hand corner of the main menu), and click on the Fetch/Validate link next to the ORCID field. This will take you to the ORCID site and allow you to create a new iD or authenticate a pre-existing iD in Editorial Manager. Please see the following video for instructions on linking an ORCID iD to your Editorial Manager account: https://www.youtube.com/watch?v=_xcclfuvtxQ

7. Please include your tables as part of your main manuscript and remove the individual files. Please note that supplementary tables (should remain/ be uploaded) as separate "supporting information" files.

8. Thank you for stating the following in the Funding Section of your manuscript:

(This study was supported by (the Chengdu Science and Technology

Bureau (No.2017FZ0058)) and (the Health Department of Sichuan Province

(No.17PJ474)))

We note that you have provided funding information that is not currently declared in your Funding Statement. However, funding information should not appear in the Acknowledgments section or other areas of your manuscript. We will only publish funding information present in the Funding Statement section of the online submission form.

Please remove any funding-related text from the manuscript and let us know how you would like to update your Funding Statement. Currently, your Funding Statement reads as follows:

(No)

Please include your amended statements within your cover letter; we will change the online submission form on your behalf.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Yes

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: Thank you for the opportunity to review this study.

The authors have conducted a retrospective observational study looking for an association between hematocrit and 30 day mortality in a septic population. The method proposed is a multivariate model of a single center retrospective dataset (MIMIC IV). They found a correlation between low HCT value at admission and increased 30d mortality.

The study finding are not very original. I have the following concerns and comments:

1) Introduction paragraph (line 81-93) seems repeated

2) reporting checklist is missing, I suggest to use the RECORD guidelines (The REporting of studies Conducted using Observational Routinely-collected health Data ) Please refer to http://journals.plos.org/plosmedicine/article?id=10.1371/journal.pmed.1001885

3) Please indicate in detail how do you selected the sepsis population in the MIMIC database. Did you used the "concepts" script for mimic-iv in order to improve the reproducibility of the analysis? (Johnson, A. E., Stone, D. J., Celi, L. A., & Pollard, T. J. (2018). The MIMIC Code Repository: enabling reproducibility in critical care research. Journal of the American Medical Informatics Association : JAMIA, 25(1), 32–39. or an extraction based on ICD-9 codes (Epidemiology of severe sepsis in the United States: analysis of incidence, outcome, and associated costs of care. Angus DC, Linde-Zwirble WT, Lidicker J, Clermont G, Carcillo J, Pinsky MRCrit Care Med. 2001 Jul; 29(7):1303-10.) and Martin et al. (The epidemiology of sepsis in the United States from 1979 through 2000. Martin GS, Mannino DM, Eaton S, Moss M

N Engl J Med. 2003 Apr 17; 348(16):1546-54)

4) Line 119-120 report the use of non parametric test, please specify the distribution of data

Reviewer #2: Comments for authors:

Dear authors,

You have presented a large-scale retrospective analysis on the association between hematocrit and mortality in patients with sepsis.

The topic is interesting and it would be important to find a real contribution in this setting.

Please give your statement to the following points:

1. Abstract

The abstract is clear enough.

2. Introduction

- The introduction was repeated twice; delete the second repetition.

3. Materials and Methods

- It has not been included if a sample size estimation has been carried out; has it been calculated?

- It is not clear enough to me whether the selected patients were already diagnosed with sepsis upon entering the ICU.

4.Results

- In the figure 1, please add legend.

5. Discussion

- It is written that "The HCT levels obtained in our research were all information of patients with sepsis on the first day after entering the ICU, which suppressed the progression of the disease and changes in laboratory parameter values induced by fluid resuscitation therapy"; but the patients who had already been diagnosed with sepsis upon admission to the ICU could not have already started fluid resuscitation therapy? Or could not they have been more fragile patients already in fluid overload?

- Please specify better the clinical message that the authors want to send.

6. Conclusion

- In my opinion the words "hematocrit during ICU" is misleading; better to specify “hematocrit in the first 24 hours after ICU admission”.

7. Tables

- Table 2 appears to be repeated; please check the data.

- Tables 1 and 2 lack legends; please enter.

8. References

Please check the journal’s guidelines

It would be necessary to revisit the English language.

Best regards

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: Yes: Alberto Noto

Reviewer #2: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

Submitted filename: review PLOS-ONE 1.doc

Click here for additional data file.

10 Feb 2022

Dear editors and reviewers:

Thanks for your letter and the reviewers' comments about our manuscript entitled “Association between hematocrit and the 30-day mortality of patients with sepsis: a retrospective analysis based on the large-scale clinical database MIMIC-IV”.All these suggestions are valuable for me to revise my manuscript. We have studied comments carefully and have made corrections which we hope meet with approval.

Suggestions from editors:

The topic is interesting and it would be important to find a real contribution in this setting.

Please give your statement to the following points:

1. Abstract

The abstract is clear enough.

2. Introduction

- The introduction was repeated twice; delete the second repetition.

3. Materials and Methods

- It has not been included if a sample size estimation has been carried out; has it been calculated?

- It is not clear enough to me whether the selected patients were already diagnosed with sepsis upon entering the ICU.

4.Results

- In the figure 1, please add legend.

5. Discussion

- It is written that "The HCT levels obtained in our research were all information of patients with sepsis on the first day after entering the ICU, which suppressed the progression of the disease and changes in laboratory parameter values induced by fluid resuscitation therapy"; but the patients who had already been diagnosed with sepsis upon admission to the ICU could not have already started fluid resuscitation therapy? Or could not they have been more fragile patients already in fluid overload?

- Please specify better the clinical message that the authors want to send.

6. Conclusion

- In my opinion the words "hematocrit during ICU" is misleading; better to specify “hematocrit in the first 24 hours after ICU admission”.

7. Tables

- Table 2 appears to be repeated; please check the data.

- Tables 1 and 2 lack legends; please enter.

8. References

Please check the journal’s guidelines

It would be necessary to revisit the English language.

Dear editors,the revised portion are marked in red in our paper.

Responses to Reviwers:

Reviewer #1

1. -Introduction paragraph (line 81-93) seems repeated

We are very sorry for our neligence of this paragraph.And then the repetitive sentence in the introductory paragraph has been deleted.

2. -reporting checklist is missing, I suggest to use the RECORD guidelines (The REporting of studies Conducted using Observational Routinely-collected health Data ) Please refer to http://journals.plos.org/plosmedicine/article?id=10.1371/journal.pmed.100188-

Considering the reviwer’s suggestion,we have made some modifications based on the guidelines.

3.-Please indicate in detail how do you selected the sepsis population in the MIMIC database. Did you used the "concepts" script for mimic-iv in order to improve the reproducibility of the analysis? (Johnson, A. E., Stone, D. J., Celi, L. A., & Pollard, T. J. (2018). The MIMIC Code Repository: enabling reproducibility in critical care research. Journal of the American Medical Informatics Association : JAMIA, 25(1), 32–39. or an extraction based on ICD-9 codes (Epidemiology of severe sepsis in the United States: analysis of incidence, outcome, and associated costs of care. Angus DC, Linde-Zwirble WT, Lidicker J, Clermont G, Carcillo J, Pinsky MRCrit Care Med. 2001 Jul; 29(7):1303-10.) and Martin et al. (The epidemiology of sepsis in the United States from 1979 through 2000. Martin GS, Mannino DM, Eaton S, Moss M

N Engl J Med. 2003 Apr 17; 348(16):1546-54)

Firstly,we download the zip package of the database from the official website for a series of installation, and then enter the interface of PostgreSQL 6.0. We use SELECT*FROM statement to find the icd-9 code of sepsis in d_icd_diagnoses, and then use SELECT WHERE statement to find the related icd-9 code of sepsis in diagnoses_icd (99591/99592)We used the "concepts" script for mimic-iv in order to improve the reproducibility of the analysis.

4.-Line 119-120 report the use of non parametric test, please specify the distribution of data

Considering the reviewer’s suggestion the data distribution has been shown in Table 1.

Reviewer #2

1.Introduction

- The introduction was repeated twice; delete the second repetition.

Thank you for your careful review.The repetitive sentence in the introductory paragraph has been deleted.

2.Materials and Methods

- It has not been included if a sample size estimation has been carried out; has it been calculated?

- It is not clear enough to me whether the selected patients were already diagnosed with sepsis upon entering the ICU.

This sample size has been calculated.The data of sepsis patients we present in the MIMIC IV database by SQL statements are diagnosed on admission to the ICU.This can be determined by referring to the r

elevant details of this database on the official website.(https://physionet.org/content/mimiciv/0.4/).

3.Results

- In the figure 1, please add legend.

Thank you for your careful review.We have added the legend.(Details can be found in the revised version)

4.Discussion

- It is written that "The HCT levels obtained in our research were all information of patients with sepsis on the first day after entering the ICU, which suppressed the progression of the disease and changes in laboratory parameter values induced by fluid resuscitation therapy"; but the patients who had already been diagnosed with sepsis upon admission to the ICU could not have already started fluid resuscitation therapy? Or could not they have been more fragile patients already in fluid overload?

Thank you for the reviewer's suggestion, We have reworked it and marked in red.

5.Conclusion

- In my opinion the words "hematocrit during ICU" is misleading; better to specify “hematocrit in the first 24 hours after ICU admission”.

Thank you for the reviewer's suggestion, I have adopted your revision and made the statement changes.

6.Tables

- Table 2 appears to be repeated; please check the data.

- Tables 1 and 2 lack legends; please enter.

The table has been made in detail。（Details can be found in the revised version）

7.References

Please check the journal’s guidelines

It would be necessary to revisit the English language.

Changes have been made to the reference specification.

Thank you for your careful review. We really appreciate your efforts in reviewing our manuscript during this unprecedented and challenging time. We wish good health to you, your family, and community. Your careful review has helped to make our study clearer and more comprehensive.

Submitted filename: Response to Reviewers.docx

Click here for additional data file.

8 Mar 2022

Association between hematocrit and the 30-day mortality of patients with sepsis: a retrospective analysis based on the large-scale clinical database MIMIC-IV

PONE-D-21-35992R1

Dear Dr. luo,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Andrea Cortegiani, M.D.

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: (No Response)

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: Yes: Alberto Noto

15 Mar 2022

PONE-D-21-35992R1

Association between hematocrit and the 30-day mortality of patients with sepsis: a retrospective analysis based on the large-scale clinical database MIMIC-IV

Dear Dr. luo:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Andrea Cortegiani

Academic Editor

PLOS ONE