Skip Navigation
Skip to contents

ACC : Acute and Critical Care

OPEN ACCESS
SEARCH
Search

Articles

Page Path
HOME > Acute Crit Care > Volume 39(3); 2024 > Article
Original Article
Pulmonary
Increased red cell distribution width predicts mortality in COVID-19 patients admitted to a Dutch intensive care unit
Anthony D. Mompiere1orcid, Jos L.M.L. le Noble1,2orcid, Manon Fleuren-Janssen1orcid, Kelly Broen3orcid, Frits van Osch4,5orcid, Norbert Foudraine1orcid
Acute and Critical Care 2024;39(3):359-368.
DOI: https://doi.org/10.4266/acc.2023.01137
Published online: August 22, 2024

1Department of Intensive Care, VieCuri Medical Center, Venlo, the Netherlands

2Department of Pharmacology and Toxicology, Maastricht University, Maastricht, the Netherlands

3Department of Clinical Chemistry and Hematology, VieCuri Medical Center, Venlo, the Netherlands

4Department of Clinical Epidemiology, VieCuri Medical Center, Venlo, the Netherlands

5Department of Epidemiology, NUTRIM School of Nutrition and Translational Research in Metabolism, Maastricht University, Maastricht, the Netherlands

Corresponding author: Anthony D. Mompiere Department of Intensive Care, VieCuri Medical Center, Venlo Tegelseweg 210, 5900 BX, Venlo-NL, the Netherlands Tel: +352-621-717-588 E-mail: a.mompiere@student.maastrichtuniversity.nl
• Received: September 1, 2023   • Revised: June 18, 2024   • Accepted: July 29, 2024

© 2024 The Korean Society of Critical Care Medicine

This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/4.0/) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.

  • 1,291 Views
  • 172 Download
prev next
  • Background
    Abnormal red blood cell distribution width (RDW) is associated with poor cardiovascular, respiratory, and coronavirus disease 2019 (COVID-19) outcomes. However, whether RDW provides prognostic insights regarding COVID-19 patients admitted to the intensive care unit (ICU) was unknown. Here, we retrospectively investigated the association of RDW with 30-day and 90-day mortalities, duration of mechanical ventilation, and length of ICU and hospital stay in patients with COVID-19.
  • Methods
    This study included 321 patients with COVID-19 aged >18 years who were admitted to the ICU between March 2020 and July 2022. The outcomes were mortality, duration of mechanical ventilation, and length of stay. RDW >14.5% was assessed in blood samples within 24 hours of admission.
  • Results
    The mortality rate was 30.5%. Multivariable Cox regression analysis showed an association between increased RDW and 30-day mortality (hazard ratio [HR], 3.64; 95% CI, 1.54–8.65), 90-day mortality (HR, 3.66; 95% CI, 1.59–8.40), and shorter duration of invasive ventilation (2.7 ventilator-free days, P=0.033).
  • Conclusions
    Increased RDW in COVID-19 patients at ICU admission was associated with increased 30-day and 90-day mortalities, and shorter duration of invasive ventilation. Thus, RDW can be used as a surrogate biomarker for clinical outcomes in COVID-19 patients admitted to the ICU.
Coronavirus disease 2019 (COVID-19) is caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). Severe forms of the disease are characterized by hyperinflammation secondary to an inflated immune response that is characterized by hypotension, hypoxemia, and multiple system organ failure. Laboratory assessments of these severe forms reflect the immune response characteristic of the cytokine storm syndrome. These cytokines affect all hematological cell lineages. Many studies have suggested that hyperinflammation secondary to an inflated immune response is the primary factor responsible for the pathogenesis of COVID-19, which contributes to death.
Red blood cell distribution width (RDW) indicates the variability in the size of the circulating red blood cells (RBCs), and is a standard component of a complete blood cell count [1]. RDW is the coefficient of variation in RBC volume, or the standard deviation (SD) in RBC volume divided by the mean. An increase in RDW therefore corresponds to a decrease in mean RBC volume, increase in RBC volume variance, or both. Erythrocyte variability depends on both red cell turnover and blood transfusions [2]. RDW increases in several inflammatory states like heart disease, sepsis, chronic obstructive pulmonary disease, and acute respiratory distress syndrome (ARDS). In these clinical cases, RDW was evaluated as a new prognostic biomarker for assessing the risk of a poor outcome [3-15].
Interestingly, COVID-19 patients exhibit high RDW. Several possible scenarios can be put forward to explain this observation; these include hypoxemia, which induces erythropoietin release and increases RBC formation and RBC volume, and the release of inflammatory cytokines due to either delayed erythrocyte maturation or overstimulation of the bone marrow following SARS-CoV-2 infection, resulting in a broad range of RBC size. However, the use of increased RDW as a new prognostic biomarker for mortality has rarely been evaluated in a large number of COVID-19 patients admitted to the intensive care unit (ICU).
Based on the results of previous studies, we hypothesized that increased RDW is associated with in-hospital mortality and increased length of stay in this population. Hence, in this study, we aimed to retrospectively investigate the association of RDW with mortality (30-day and 90-day periods), duration of mechanical ventilation, and length of ICU and hospital stay in critically ill patients with COVID-19. The results will significantly impact the rationale for the clinical use of RDW as a surrogate biomarker to predict patient outcomes.
Ethical Approval
The approval by the Medical Ethics Review Committee of Maastricht University Medical Centre was obtained (No. 2023-3656). Written informed consent was waived in accordance with the affiliation mentioned.
Study Design and Patient Selection (Inclusion and Exclusion Criteria)
This single-center retrospective cohort study was performed at the ICU. Patients aged >18 years with COVID-19 viral pneumonia confirmed using reverse transcription-polymerase chain reaction (RT-PCR) and/or radiology imaging upon admission to the ICU between March 1, 2020, and July 31, 2022, were included. Data were retrieved from the ICU’s electronic medical record (Mediscore) and the digital patient database and management system (HIX, ChipSoft). Patient registries were consulted to exclude patients <18 years of age, readmitted patients, and patients who underwent erythrocyte transfusions.
Data Collection
Data regarding the following variables were extracted: sex, age, mortality status, method of COVID-19 diagnosis, vaccination rates, Karnofsky performance scales, body mass index (BMI; kg/m2), Acute Physiology and Chronic Health Evaluation (APACHE) II scores, APACHE IV scores, laboratory values, dates, and methods of mechanical ventilation [16].
Measurement Technique and Laboratory Analyses Using Sysmex XN-9000
Blood was collected in EDTA tubes. The Sysmex XN-9000 was used for obtaining the complete blood cell counts, and laboratory values at ICU admission were specifically considered; therefore, only values related to the first 24 hours were considered relevant. After 24 hours, many confounders must be considered (i.e., shock, infection, and iatrogenic complications). Consequently, laboratory values obtained after 24 hours of admission were ignored for this study. The Sysmex XN-9000 was manufactured by Sysmex Inc., and distributed by Sysmex Nederland BV in Etten-Leur.
Outcomes
The primary outcome was 30-day mortality. The secondary outcomes were 90-day mortality, length of stay in the ICU, length of stay at the hospital, and duration of mechanical ventilation. Patients who survived longer than 30 days and 90 days after admission were considered “survivors” for the respective 30-day and 90-day survival analyses.
Statistical Analysis
The final dataset was imported into SPSS Statistical software version 27.0 (IBM Corp.) for all statistical analysis. Descriptive statistics were performed to provide an overview of the characteristics of the study population. The study population was separated into high RDW (>14.5%) and low RDW (≤14.5%) groups under the assumption that increased RDW was associated with worse clinical outcomes. A 14.5% RDW cut-off was chosen based on the results of previous studies examining mortality in hospitalized patients with COVID-19 [3]. For descriptive statistics, RDW was coded as a categorical variable, whereas other laboratory values, APACHE score, and durations were considered continuous variables. An analysis of variance test coupled with a two-tailed independent t-test was used to compare sex and mortality distributions, mean age, laboratory values, APACHE scores, length of stay and duration of mechanical ventilation. A multivariable Cox regression analysis was performed to analyze the association between RDW and mortality. Survival plots were plotted and hazard ratios (HRs) with 95% CIs were obtained. Poisson regression with negative binomial distribution was performed to evaluate the length of ICU and hospital stay and duration of mechanical ventilation, for which ventilation-free duration was analyzed. Two-sided P-values < 0.05 were considered significant in all analyses. For the illustration of the length of ICU and hospital stays and duration of mechanical ventilation, RDW was analyzed both as continuous and categorical variables.
Baseline Demographics and Clinical Data
During the study period, 410 patients were admitted to the ICU with viral pneumonia. Patients with diagnoses other than COVID-19 (n=6), who had erythrocyte transfusions (n=56), were logistic ICU admissions (n=1) or readmissions (n=13), patients with missing data (n=1), or were aged <18 years (n=1) were excluded. In total, 321 patients were included and were analyzed in the study retrospectively (Figure 1). Most patients (317/321, 98.8%) had COVID-19 confirmed using RT-PCR. Four patients were included despite absence of PCR confirmation because of highly suggestive clinical presentations (i.e., dry cough, dyspnea, fever, hypoxic respiratory insufficiency, and tiredness), chest x-rays, and/or computed tomography (CT severity scores ranging from 11 to 22). The mean and SD age of the study population was 63±12 years (Table 1). Most patients (216/321, 67.3%) were male. Ninety-eight patients (30.5%) had deceased at the end of this study. Most patients (245/321, 76.3%) showed low RDW levels at ICU admission (≤14.5%). The mean and SD RDW in this study population was 13.8%±1.51%, as shown in Figure 2. The distribution was positively skewed, with a median RDW of 13.5%. There were no differences in vaccination rates or Karnofsky performance scales between the populations. The high RDW population was more obese than the low RDW population (BMI: 30.1±5.8 kg/m2 vs. 32.5±10.3 kg/m2).
30-Day Mortality
The 30-day mortality rate in patients with a high RDW (28/76, 36.8%) was higher than that in patients with a low RDW (60/245, 24.5%). The unadjusted 30-day mortality in patients with high RDW was significantly higher than that in patients with low RDW (hazard ratio [HR], 1.80; 95% CI, 1.15–2.82). The adjustment for all baseline characteristics scores revealed an even greater association with 30-day mortality (HR, 3.64; 95% CI, 1.54–8.65), as reported in Table 2. Additionally, Cox proportional hazards analysis by RDW quartiles showed associations with 30-day mortality. Before adjustment for baseline characteristics, the top quartile group (RDW >14.4%) was associated with a 30-day mortality risk (HR, 2.92; 95% CI, 1.54–5.56). In this subset, the top RDW quartile showed an even higher association with 30-day mortality than the reference when adjusted for all mentioned baseline characteristics (HR, 3.39; 95% CI, 1.09–10.58). The cumulative survival plot in Figure 3 illustrates the significant impact of RDW on 30-day mortality.
90-Day Mortality
Patients in the high RDW group showed a 90-day mortality rate of 42.1% (32/76) compared to 26.1% (64/245) in the low RDW population. The unadjusted survival analysis calculated 90-day mortality in the Cox proportional hazards model of the univariable survival analysis for this population resulted in an HR of 1.93 (95% CI, 1.26–2.95), as reported in Table 2. Similar to 30-day mortality, the correction for all mentioned baseline characteristics revealed even higher association with 90-day mortality (HR, 3.66; 95% CI, 1.59–8.40), as reported in Table 2. The RDW quartiles showed a gradual increase in 90-day mortality risks with increase in RDW, regardless of adjustment. The top quartile group showed an almost threefold unadjusted 90-day mortality risk compared to the patient group with RDW <12.9% (HR, 2.94; 95% CI, 1.61–5.35). In this subset, the top quartile showed an even greater association with 90-day mortality when adjusted for all baseline characteristics, including APACHE-IV scores and BMI (HR, 3.79; 95% CI, 1.24–11.56).
Length of Stay
The mean length of stay in the ICU was 7.4±7.4 days in the high RDW population and 9.4±11.8 days in the low RDW population. Similarly, the mean length of stay in the hospital was 12.2±12.5 days in the high RDW population and 14.8±14.7 days in the low RDW population. High RDW was not significantly associated with shorter length of ICU stay (P=0.074) or hospital stay (P=0.134). These results are illustrated in Figure 4.
Duration of Mechanical Ventilation
A High RDW was associated with shorter duration of mechanical ventilation method. The results were unaltered with and without correction for overall mortality. The difference in ventilator-free days between high and low RDW populations was 2.7 days for invasive ventilation (P=0.033). When corrected for BMI, no statistical difference between both study groups was demonstrated (2.4 days, P=0.07) (Figure 4).
We retrospectively investigated the association of RDW with 30-day and 90-day mortalities, duration of mechanical ventilation, and length of ICU and hospital stay in patients with COVID-19. The main finding of our study is that an increase in RDW was associated with an increased mortality, independent of age and APACHE-IV scores, BMI, mean corpuscular volume, hematocrit, hemoglobin, ferritin, and C-reactive protein level. No differences in ICU and hospital lengths of stay were found, indicating that RDW does not affect the length of stay. The correction for mortality did not affect the length of stay at the ICU, implying that lead time bias could be considered negligible. We also demonstrated a shorter duration of mechanical ventilation in patients with a high RDW.
Our results imply that RDW may act as a potential new surrogate biomarker for COVID-19-related mortality in ICU patients. These results are in agreement with those observed in several cohorts of patients with ARDS, RDW of whom had increased [5,7,11,13]. Our study contributes to the field of biomarker research as it shows that RDW is associated with significant mortality. Age and APACHE IV scores were confounders in the multivariable Cox regression analysis, as they are potent predictors of mortality in ICU patients. Foy et al. [3] stratified hospital mortality in hospitalized COVID-19 patients by categorical age, obtaining separate proportional Cox hazards per age group. The impact of age was not investigated further to avoid distraction from the outcomes chosen for this study.
In our study, despite some small expected loss of power, the approach of using RDW in quartiles revealed changes in mortality with progressive increase in RDW, which is consistent with the observations of studies on ARDS and non-ICU COVID-19-positive patients [8,17,18]. Hornick et al. [19] showed that increasing RDW quartiles were also associated with mortality in patients with COVID-19. We hypothesized that hypoxemia due to severe COVID-19, via erythropoiesis, increases RDW. This implies a relationship between either severity of COVID-19 or duration of COVID-19-associated hypoxemia and increase in RDW. In this case, patients admitted to the ICU should reflect more anisocytosis than non-ICU patients, although this is unclear at present. ICU mortality might increase after admission from the emergency department in comparison to admission from the general ward because of absence of oxygen therapy in the former [4,20,21].
The relation between increased RDW and mortality in COVID-19 patients has now been extended from general ward patients to ICU patients. This indicates that COVID-19 infection may lead to unfavorable outcomes when accompanied by RDW increment. The mechanism via which COVID-19 may lead to such RDW increments in some patients is not currently understood and requires further investigations. It is difficult to explain the shorter invasive ventilation duration, but upon correction it might be related to the obesity paradox as reported in previous literature [22-26]. However, following adjustment for BMI we were unable to demonstrate that BMI may be protective.
Our study is novel and characterizes the benefits of using RDW in ICU patients with COVID-19. The inclusion of an exclusively ICU-admitted patient population with confirmed COVID-19 is one of the strengths of this study, as only few studies have investigated this association [3,15,27]. Second, a large sample size was used in this study that ensured adequate power of 0.8, thereby contributing to the field of critical care medicine with the use of RDW as a potential surrogate biomarker. Insights regarding the lengths of stay at the hospital and ICU will be financially relevant for hospital and healthcare management. Our study also has some limitations. The results may have been influenced by the rapidity with which research on COVID-19 pathophysiology has progressed since the emergence of the pandemic and the frequent changes in treatment guidelines and vaccination regimens. The vaccination status in our study group was heterogenous, leading to possible selection and workup bias since the a priori mortality risks could not be considered equal [28,29]. The vaccination against COVID-19 in the Netherlands was not available prior to January 2021, but the first vaccinated patient in this study sample appeared only 3 months later [30,31]. Nevertheless, a large proportion was unvaccinated either through mistrust of the existing vaccine, or underestimation of the COVID19 pandemic.
It's also noticeable that both populations manifested high Karnofsky performance scales, leading to potential selection bias. The Dutch healthcare system seems to have a stricter approach towards hospital and ICU admissions, due to bed and personnel capacity issues [32]. Patients with high clinical frailty scores are often poor candidates for ICU admission because of their prolonged expected stay. In addition, a large range of diagnostics and treatment regimens was experimented with during these 29 months which could have resulted in some form of selection or workup bias [33-35].
The pathophysiological mechanisms underlying the increase in RDW in COVID-19 patients remain unknown. Certain pathophysiological mechanisms (for example, inflammation) causing RDW elevation have been described in the context of certain cardiovascular diseases [1]. Inflammation and bone marrow hyperactivation will subsequently lead to the release of immature cells into the circulation. Besides cells of the white blood cell lineage, immature red cells are found to be involved in the activation of severe inflammation [36]. We speculated that RDW elevation in critically ill COVID-19 patients reflects a hyperinflammatory state. Markers of cytokine storm syndrome (i.e., D-dimer and ferritin) were discovered during the pandemic and were therefore not assessed in every patient. Investigation regarding elevation in RDW was interesting, as existing literature highlights the potential clinical relevance of gradual RDW elevations [10]. However, ICU admissions are associated with many confounding interventions that may impact hematological variables. A deeper understanding regarding the underlying relationship might affect clinical decision-making in the future.
Finally, the use of RDW as a crucial marker for estimating the severity of COVID-19 can be justified. Increased RDW can be included in severity algorithms (for example, the COVID Inpatient Risk Calculator score, Severe COVID-19 Adaptive Risk Predictor) [37]. The potentially confounding impact of evolving screening and management tools on the association of increased RDW with COVID-19 mortality could be worthy of investigation in a larger study. RDW is a promising predictive biomarker for mortality in COVID-19 patients admitted to the ICU. The findings provide valuable insights for clinically relevant outcomes in this patient population. Further studies are needed to explore the underlying pathophysiology.
▪ High red blood cell distribution width (RDW) in intensive care unit (ICU) coronavirus disease 2019 (COVID-19) patients was associated with increased 30-day and 90-day mortality.
▪ High RDW in ICU COVID-19 patients required shorter period of invasive ventilation.
▪ RDW can be a surrogate biomarker for clinical outcomes in ICU COVID-19 patients.

CONFLICT OF INTEREST

No potential conflict of interest relevant to this article was reported.

FUNDING

None.

AUTHOR CONTRIBUTIONS

Conceptualization: AM, JLMLN, MFJ, KB, NF. Methodology: AM, JLMLN, MFJ, KB, NF. Formal analysis: AM, JLMLN, MFJ, FO, NF. Data curation: AM, JLMLN, MFJ, KB, NF. Visualization: AM, JLMLN, MFJ, FO, NF. Project administration: AM, JLMLN, MFJ, FO, NF. Writing - original draft: all authors. Writing - review & editing: all authors. All authors read and agreed to the published version of the manuscript.

None.
Figure 1.
Patient selection with elaboration of exclusion criteria. After the exclusion of 89 patients, 321 patients were included in the analysis. Seventy-six of these patients presented with increased red blood cell distribution width (RDW), while 245 patients did not. ICU: intensive care unit; COVID-19: coronavirus disease 2019.
acc-2023-01137f1.jpg
Figure 2.
Distribution of red blood cell distribution width (RDW) values in the study population. The mean RDW was 13.8% and the median RDW was 13.5%. The minimum RDW was 11.4% and the maximum was 21.2%. The standard deviation was 1.51.
acc-2023-01137f2.jpg
Figure 3.
Mortality plot by red blood cell distribution width (RDW) quartile groups with cumulative survival time in days. The cumulative survival plot shows the progressively increasing mortality with increased RDW in coronavirus disease 2019 (COVID-19) patients when separated in quartiles. It also shows a significant 30-day mortality, which stabilises when 90-day mortality is investigated. The RDW populations of this survival analysis were separated in quartiles.
acc-2023-01137f3.jpg
Figure 4.
Scatter plots showing the trends of distributions in length of stay (LOS) and durations of ventilation, with continuous red blood cell distribution width (RDW) levels on the x-axis and either LOS or durations in days on the y-axis. The increased RDW group (>14.5%, red group) has shorter LOS at the intensive care unit (ICU; A) and at the hospital (B) on average, as well as reduced durations of invasive (C) and prone ventilation (D) compared to those of the normal RDW group.
acc-2023-01137f4.jpg
Table 1.
Descriptive statistics of baseline characteristics (durations and laboratory values) by RDW cut-off of 14.5%
Variable RDW ≤14.5% (n=245) RDW >14.5% (n=76) P-value
Age (yr) 63±12 65±12 0.135
APACHE II score 15.0±5.0 16.0±5.6 0.149
APACHE IV score 57.0±18.8 62.2±20.7 0.041
Body mass index (kg/m2) 30.1±5.8 32.5±10.3 0.008
Vaccination rate (%) 13.9±0.35 15.8±0.37 0.709
Median KPS 100±14.4 (30–100) 100±14.2 (40–100) 0.129
Mortality (%) 26.0±44.0 45±50.1 0.004
Male sex (%) 69.0±46.4 61.8±48.9 0.263
Length of stay & mechanical ventilation
 ICU stay (day) 9.4±11.8 7.4±7.4 0.086
 Hospital stay (day) 14.8±14.7 12.2±12.5 0.081
 Prone ventilation (day) 2.6±4.5 1.4±1.0 0.223
 Invasive ventilation (day) 9.6±11.5 6.9±7.1 0.042
 CPAP (day) 3.5±4.9 1.0±6.0 0.108
 Optiflow (day) 1.9±1.2 3.0±3.4 0.570
 Intubation (day) 1.6±4.3 1.1±2.5 0.516
Laboratory value
 C-reactive protein (mg/ml) 132.9±96.4 131.4±93.1 0.292
 Hemoglobin (mmol/L) 8.1±1.0 7.9±1.2 <0.001
 Hematocrit (L/L) 0.4±0.0 0.4±0.1 0.017
 Mean corpuscular volume (fmol) 89.7±5.2 89.7±5.5 0.100
 Mean corpuscular hemoglobin (fL) 1,878.0±115.5 1,846.63±136.8 <0.001
 Leukocytes (×109/L) 9.9±4.5 9.8±4.3 0.302
 Thrombocytes (×109/L) 253.0±97.7 231.9±90.4 0.775
 Neutrophilic granulocytes (×109/L) 8.4±4.0 8.6±5.2 0.550
 Lymphocytes (×109/L) 0.9±0.4 0.8±0.5 0.468
 Monocytes (×109/L) 0.6±0.8 0.5±0.2 0.844
 Prothrombin time (INR) 1.1±0.3 1.2±0.3 0.305
 APTT (sec) 25.6±8.0 26.0±6.4 0.681
 Fibrinogen (g/L) 6.0±1.7 5.6±1.8 0.955
 D-dimer (μg/L) 4,842.5±8,862.1 7,901.9±10,785.7 0.096
 Ferritin (μg/L) 2,057.3±2,282.7 1,202.5±1,005.9 0.005

Values are presented as mean±standard deviation (SD) or mean±SD (range).

RDW: red cell distribution width; APACHE: Acute Physiology and Chronic Health Evaluation; KPS: Karnofsky performance scale; ICU: intensive care unit; CPAP: continuous positive airway pressure; INR: International normalized ratio; APTT: activated plasminogen thromboplastin time.

Table 2.
Cox proportional unadjusted hazards for 30-day and 90-day mortalities, and proportional adjusted hazards for age, BMI, sex, CRP, MCH, Hb, Ht, ferritin, and APACHE IV scores
RDW population 30-Day
90-Day
HR (95% CI) aHR (95% CI) HR (95% CI) aHR (95% CI)
RDW >14.5% vs. ≤14.5% 1.80 (1.15–2.82) (P=0.010) 3.64 (1.54–8.65) (P=0.003) 1.93 (1.26–2.95) (P=0.002) 3.66 (1.59–8.40) (P=0.002)
RDW quartile 0 Reference Reference Reference Reference
RDW quartile 1 2.09 (1.08–4.04) (P=0.028) 1.31 (0.42–4.11) (P=0.647) 1.92 (1.03–3.60) (P=0.042) 1.40 (0.45–4.36) (P=0.559)
RDW quartile 2 1.92 (0.98–3.75) (P=0.057) 0.67 (0.17–2.62) (P=0.566) 1.76 (0.93–3.33) (P=0.084) 0.86 (0.24–3.09) (P=0.814)
RDW quartile 3 2.92 (1.54–5.56) (P=0.001) 3.39 (1.09–10.58) (P=0.036) 2.94 (1.61–5.35) (P<0.001) 3.79 (1.24–11.56) (P=0.019)

Adjusted hazard ratios (aHR) were dependent on age, body mass index (BMI), sex, C-reactive protein (CRP), mean corpuscular hemoglobin (MCH), hemoglobin (Hb), hematocrit (Ht), ferritin, and Acute Physiology and Chronic Health Evaluation (APACHE) IV score. These mortality endpoints were also evaluated by quartile groups, with quartile group 0 (red blood cell distribution width [RDW] <12.9%) acting as a reference. Quartile group 1: 12.9%<RDW<13.5%, Quartile group 2: 13.5%<RDW<14.4%, Quartile group 3: RDW >14.4%. Quartile 3 shows increased proportional hazards compared to quartile 1, revealing an increased association with mortality.

  • 1. Li N, Zhou H, Tang Q. Red blood cell distribution width: a novel predictive indicator for cardiovascular and cerebrovascular diseases. Dis Markers 2017;2017:7089493. ArticlePubMedPMCPDF
  • 2. Spadaro S, Taccone FS, Fogagnolo A, Franchi F, Scolletta S, Ragazzi R, et al. The effects of blood transfusion on red blood cell distribution width in critically ill patients: a pilot study. Transfusion 2018;58:1863-9.ArticlePubMedPDF
  • 3. Foy BH, Carlson JC, Reinertsen E, Padros I Valls R, Pallares Lopez R, Palanques-Tost E, et al. Association of red blood cell distribution width with mortality risk in hospitalized adults with SARS-CoV-2 infection. JAMA Netw Open 2020;3:e2022058.ArticlePubMedPMC
  • 4. Jandaghian S, Vaezi A, Manteghinejad A, Nasirian M, Vaseghi G, Haghjooy Javanmard S. Red blood cell distribution width (RDW) as a predictor of in-hospital mortality in COVID-19 patients: a cross sectional study. Arch Acad Emerg Med 2021;9:e67.PubMedPMC
  • 5. Yu XS, Chen ZQ, Hu YF, Chen JX, Xu WW, Shu J, et al. Red blood cell distribution width is associated with mortality risk in patients with acute respiratory distress syndrome based on the Berlin definition: a propensity score matched cohort study. Heart Lung 2020;49:641-5.ArticlePubMed
  • 6. Wang AY, Ma HP, Kao WF, Tsai SH, Chang CK. Red blood cell distribution width is associated with mortality in elderly patients with sepsis. Am J Emerg Med 2018;36:949-53.ArticlePubMed
  • 7. Dankl D, Rezar R, Mamandipoor B, Zhou Z, Wernly S, Wernly B, et al. Red cell distribution width is independently associated with mortality in sepsis. Med Princ Pract 2022;31:187-94.ArticlePubMedPMCPDF
  • 8. Ghimire R, Shakya YM, Shrestha TM, Neupane RP. The utility of red cell distribution width to predict mortality of septic patients in a tertiary hospital of Nepal. BMC Emerg Med 2020;20:43. ArticlePubMedPMCPDF
  • 9. Tertemiz KC, Ozgen Alpaydin A, Sevinc C, Ellidokuz H, Acara AC, Cimrin A. Could "red cell distribution width" predict COPD severity? Rev Port Pneumol (2006) 2016;22:196-201.ArticlePubMed
  • 10. Felker GM, Allen LA, Pocock SJ, Shaw LK, McMurray JJ, Pfeffer MA, et al. Red cell distribution width as a novel prognostic marker in heart failure: data from the CHARM Program and the Duke Databank. J Am Coll Cardiol 2007;50:40-7.PubMed
  • 11. van Kimmenade RR, Mohammed AA, Uthamalingam S, van der Meer P, Felker GM, Januzzi JL. Red blood cell distribution width and 1-year mortality in acute heart failure. Eur J Heart Fail 2010;12:129-36.ArticlePubMed
  • 12. Bulut O, Akcakaya A, Bulut N, Ovali F. Elevated red cell distribution width as a useful marker in neonatal sepsis. J Pediatr Hematol Oncol 2021;43:180-5.ArticlePubMed
  • 13. Martin SL, Desai S, Nanavati R, Colah RB, Ghosh K, Mukherjee MB. Red cell distribution width and its association with mortality in neonatal sepsis. J Matern Fetal Neonatal Med 2019;32:1925-30.ArticlePubMed
  • 14. Guaní-Guerra E, Torres-Murillo B, Muñoz-Corona C, Rodríguez-Jiménez JC, Macías AE, Scavo-Montes DA, et al. Diagnostic accuracy of the RDW for predicting death in COVID-19. Medicina (Kaunas) 2022;58:613. ArticlePubMedPMC
  • 15. Lorente L, Martín MM, Argueso M, Solé-Violán J, Perez A, Marcos Y Ramos JA, et al. Association between red blood cell distribution width and mortality of COVID-19 patients. Anaesth Crit Care Pain Med 2021;40:100777. ArticlePubMed
  • 16. Silva JC, Giraldi T, Coutinho CM, Carvalho Filho MA, Fernandes DC, Santos TM. Reduced physical functional performance before hospitalization predicts life support limitations and mortality in nonsurgical intensive care unit patients. Rev Bras Ter Intensiva 2022;34:166-75.ArticlePubMedPMC
  • 17. Moghimi M, Jozpanahi M, Khodadadi K, Saeed SP, Pirsaraie SV, Jalili N. Red cell distribution width, a predictive factor in immunocompromised patients with COVID-19: a comparison retrospective study between cancer and kidney transplant patients. Eur J Transl Myol 2022;32:10582. ArticlePubMedPMCPDF
  • 18. Henry BM, Benoit JL, Benoit S, Pulvino C, Berger BA, Olivera MH, et al. Red blood cell distribution width (RDW) predicts COVID-19 severity: a prospective, observational study from the Cincinnati SARS-CoV-2 Emergency Department Cohort. Diagnostics (Basel) 2020;10:618. ArticlePubMedPMC
  • 19. Hornick A, Tashtish N, Osnard M, Shah B, Bradigan A, Albar Z, et al. Anisocytosis is associated with short-term mortality in COVID-19 and may reflect proinflammatory signature in uninfected ambulatory adults. Pathog Immun 2020;5:312-26.ArticlePubMedPMCPDF
  • 20. Kaufmann CC, Ahmed A, Brunner U, Jäger B, Aicher G, Equiluz-Bruck S, et al. Red cell distribution width upon hospital admission predicts short-term mortality in hospitalized patients with COVID-19: a single-center experience. Front Med (Lausanne) 2021;8:652707. ArticlePubMedPMC
  • 21. Sarkar S, Kannan S, Khanna P, Singh AK. Role of red blood cell distribution width, as a prognostic indicator in COVID-19: a systematic review and meta-analysis. Rev Med Virol 2022;32:e2264.ArticlePubMedPDF
  • 22. Ball L, Serpa Neto A, Pelosi P. Obesity and survival in critically ill patients with acute respiratory distress syndrome: a paradox within the paradox. Crit Care 2017;21:114. ArticlePubMedPMCPDF
  • 23. Liu QY, Chen Y, He Y, Zhu RL. Impact of obesity on outcomes in patients with acute respiratory syndrome. J Int Med Res 2021;49:3000605211024860. ArticlePubMedPDF
  • 24. Yeo HJ, Kim TH, Jang JH, Jeon K, Oh DK, Park MH, et al. Obesity paradox and functional outcomes in sepsis: a multicenter prospective study. Crit Care Med 2023;51:742-52.ArticlePubMedPMC
  • 25. Vulturar DM, Crivii CB, Orăsan OH, Palade E, Buzoianu AD, Zehan IG, et al. Obesity impact on SARS-CoV-2 infection: pros and cons "obesity paradox." A systematic review. J Clin Med 2022;11:3844. ArticlePubMedPMC
  • 26. Zhou D, Wang C, Lin Q, Li T. The obesity paradox for survivors of critically ill patients. Crit Care 2022;26:198. ArticlePubMedPMCPDF
  • 27. Zinellu A, Mangoni AA. Red blood cell distribution width, disease severity, and mortality in hospitalized patients with SARS-CoV-2 infection: a systematic review and meta-analysis. J Clin Med 2021;10:286. ArticlePubMedPMC
  • 28. Huang YZ, Kuan CC. Vaccination to reduce severe COVID-19 and mortality in COVID-19 patients: a systematic review and meta-analysis. Eur Rev Med Pharmacol Sci 2022;26:1770-6.PubMed
  • 29. Jiang Y, Rubin L, Peng T, Liu L, Xing X, Lazarovici P, et al. Cytokine storm in COVID-19: from viral infection to immune responses, diagnosis and therapy. Int J Biol Sci 2022;18:459-72.ArticlePubMedPMC
  • 30. Sanders JG, Spruijt P, van Dijk M, Elberse J, Lambooij MS, Kroese FM, et al. Understanding a national increase in COVID-19 vaccination intention, the Netherlands, November 2020-March 2021. Euro Surveill 2021;26:2100792. ArticlePubMedPMC
  • 31. Pijpers J, van Roon A, van Roekel C, Labuschagne L, Smagge B, Ferreira JA, et al. Determinants of COVID-19 vaccine uptake in the Netherlands: a nationwide registry-based study. Vaccines (Basel) 2023;11:1409. ArticlePubMedPMC
  • 32. Verweij M, van de Vathorst S, Schermer M, Willems D, de Vries M. Ethical advice for an intensive care triage protocol in the COVID-19 pandemic: lessons learned from the Netherlands. Public Health Ethics 2020;13:157-65.ArticlePubMedPMCPDF
  • 33. Tang Y, Liu J, Zhang D, Xu Z, Ji J, Wen C. Cytokine storm in COVID-19: the current evidence and treatment strategies. Front Immunol 2020;11:1708. ArticlePubMedPMC
  • 34. Siemieniuk RA, Bartoszko JJ, Zeraatkar D, Kum E, Qasim A, Martinez JP, et al. Drug treatments for covid-19: living systematic review and network meta-analysis. BMJ 2020;370:m2980. PubMed
  • 35. Prodromos C, Rumschlag T. Hydroxychloroquine is effective, and consistently so when provided early, for COVID-19: a systematic review. New Microbes New Infect 2020;38:100776. ArticlePubMedPMC
  • 36. Paulson RF, Ruan B, Hao S, Chen Y. Stress erythropoiesis is a key inflammatory response. Cells 2020;9:634. ArticlePubMedPMC
  • 37. Wongvibulsin S, Garibaldi BT, Antar AA, Wen J, Wang MC, Gupta A, et al. Development of Severe COVID-19 Adaptive Risk Predictor (SCARP), a calculator to predict severe disease or death in hospitalized patients with COVID-19. Ann Intern Med 2021;174:777-85.ArticlePubMed

Figure & Data

References

    Citations

    Citations to this article as recorded by  

      • PubReader PubReader
      • ePub LinkePub Link
      • Cite
        CITE
        export Copy
        Close
        Download Citation
        Download a citation file in RIS format that can be imported by all major citation management software, including EndNote, ProCite, RefWorks, and Reference Manager.

        Format:
        • RIS — For EndNote, ProCite, RefWorks, and most other reference management software
        • BibTeX — For JabRef, BibDesk, and other BibTeX-specific software
        Include:
        • Citation for the content below
        Increased red cell distribution width predicts mortality in COVID-19 patients admitted to a Dutch intensive care unit
        Acute Crit Care. 2024;39(3):359-368.   Published online August 22, 2024
        Close
      • XML DownloadXML Download
      Figure
      • 0
      • 1
      • 2
      • 3
      Related articles
      Increased red cell distribution width predicts mortality in COVID-19 patients admitted to a Dutch intensive care unit
      Image Image Image Image
      Figure 1. Patient selection with elaboration of exclusion criteria. After the exclusion of 89 patients, 321 patients were included in the analysis. Seventy-six of these patients presented with increased red blood cell distribution width (RDW), while 245 patients did not. ICU: intensive care unit; COVID-19: coronavirus disease 2019.
      Figure 2. Distribution of red blood cell distribution width (RDW) values in the study population. The mean RDW was 13.8% and the median RDW was 13.5%. The minimum RDW was 11.4% and the maximum was 21.2%. The standard deviation was 1.51.
      Figure 3. Mortality plot by red blood cell distribution width (RDW) quartile groups with cumulative survival time in days. The cumulative survival plot shows the progressively increasing mortality with increased RDW in coronavirus disease 2019 (COVID-19) patients when separated in quartiles. It also shows a significant 30-day mortality, which stabilises when 90-day mortality is investigated. The RDW populations of this survival analysis were separated in quartiles.
      Figure 4. Scatter plots showing the trends of distributions in length of stay (LOS) and durations of ventilation, with continuous red blood cell distribution width (RDW) levels on the x-axis and either LOS or durations in days on the y-axis. The increased RDW group (>14.5%, red group) has shorter LOS at the intensive care unit (ICU; A) and at the hospital (B) on average, as well as reduced durations of invasive (C) and prone ventilation (D) compared to those of the normal RDW group.
      Increased red cell distribution width predicts mortality in COVID-19 patients admitted to a Dutch intensive care unit
      Variable RDW ≤14.5% (n=245) RDW >14.5% (n=76) P-value
      Age (yr) 63±12 65±12 0.135
      APACHE II score 15.0±5.0 16.0±5.6 0.149
      APACHE IV score 57.0±18.8 62.2±20.7 0.041
      Body mass index (kg/m2) 30.1±5.8 32.5±10.3 0.008
      Vaccination rate (%) 13.9±0.35 15.8±0.37 0.709
      Median KPS 100±14.4 (30–100) 100±14.2 (40–100) 0.129
      Mortality (%) 26.0±44.0 45±50.1 0.004
      Male sex (%) 69.0±46.4 61.8±48.9 0.263
      Length of stay & mechanical ventilation
       ICU stay (day) 9.4±11.8 7.4±7.4 0.086
       Hospital stay (day) 14.8±14.7 12.2±12.5 0.081
       Prone ventilation (day) 2.6±4.5 1.4±1.0 0.223
       Invasive ventilation (day) 9.6±11.5 6.9±7.1 0.042
       CPAP (day) 3.5±4.9 1.0±6.0 0.108
       Optiflow (day) 1.9±1.2 3.0±3.4 0.570
       Intubation (day) 1.6±4.3 1.1±2.5 0.516
      Laboratory value
       C-reactive protein (mg/ml) 132.9±96.4 131.4±93.1 0.292
       Hemoglobin (mmol/L) 8.1±1.0 7.9±1.2 <0.001
       Hematocrit (L/L) 0.4±0.0 0.4±0.1 0.017
       Mean corpuscular volume (fmol) 89.7±5.2 89.7±5.5 0.100
       Mean corpuscular hemoglobin (fL) 1,878.0±115.5 1,846.63±136.8 <0.001
       Leukocytes (×109/L) 9.9±4.5 9.8±4.3 0.302
       Thrombocytes (×109/L) 253.0±97.7 231.9±90.4 0.775
       Neutrophilic granulocytes (×109/L) 8.4±4.0 8.6±5.2 0.550
       Lymphocytes (×109/L) 0.9±0.4 0.8±0.5 0.468
       Monocytes (×109/L) 0.6±0.8 0.5±0.2 0.844
       Prothrombin time (INR) 1.1±0.3 1.2±0.3 0.305
       APTT (sec) 25.6±8.0 26.0±6.4 0.681
       Fibrinogen (g/L) 6.0±1.7 5.6±1.8 0.955
       D-dimer (μg/L) 4,842.5±8,862.1 7,901.9±10,785.7 0.096
       Ferritin (μg/L) 2,057.3±2,282.7 1,202.5±1,005.9 0.005
      RDW population 30-Day
      90-Day
      HR (95% CI) aHR (95% CI) HR (95% CI) aHR (95% CI)
      RDW >14.5% vs. ≤14.5% 1.80 (1.15–2.82) (P=0.010) 3.64 (1.54–8.65) (P=0.003) 1.93 (1.26–2.95) (P=0.002) 3.66 (1.59–8.40) (P=0.002)
      RDW quartile 0 Reference Reference Reference Reference
      RDW quartile 1 2.09 (1.08–4.04) (P=0.028) 1.31 (0.42–4.11) (P=0.647) 1.92 (1.03–3.60) (P=0.042) 1.40 (0.45–4.36) (P=0.559)
      RDW quartile 2 1.92 (0.98–3.75) (P=0.057) 0.67 (0.17–2.62) (P=0.566) 1.76 (0.93–3.33) (P=0.084) 0.86 (0.24–3.09) (P=0.814)
      RDW quartile 3 2.92 (1.54–5.56) (P=0.001) 3.39 (1.09–10.58) (P=0.036) 2.94 (1.61–5.35) (P<0.001) 3.79 (1.24–11.56) (P=0.019)
      Table 1. Descriptive statistics of baseline characteristics (durations and laboratory values) by RDW cut-off of 14.5%

      Values are presented as mean±standard deviation (SD) or mean±SD (range).

      RDW: red cell distribution width; APACHE: Acute Physiology and Chronic Health Evaluation; KPS: Karnofsky performance scale; ICU: intensive care unit; CPAP: continuous positive airway pressure; INR: International normalized ratio; APTT: activated plasminogen thromboplastin time.

      Table 2. Cox proportional unadjusted hazards for 30-day and 90-day mortalities, and proportional adjusted hazards for age, BMI, sex, CRP, MCH, Hb, Ht, ferritin, and APACHE IV scores

      Adjusted hazard ratios (aHR) were dependent on age, body mass index (BMI), sex, C-reactive protein (CRP), mean corpuscular hemoglobin (MCH), hemoglobin (Hb), hematocrit (Ht), ferritin, and Acute Physiology and Chronic Health Evaluation (APACHE) IV score. These mortality endpoints were also evaluated by quartile groups, with quartile group 0 (red blood cell distribution width [RDW] <12.9%) acting as a reference. Quartile group 1: 12.9%<RDW<13.5%, Quartile group 2: 13.5%<RDW<14.4%, Quartile group 3: RDW >14.4%. Quartile 3 shows increased proportional hazards compared to quartile 1, revealing an increased association with mortality.


      ACC : Acute and Critical Care
      TOP