a b s t r a c t

Introduction: RDW is a Prognostic biomarker and associated with mortality in cardiovascular disease, stroke and metabolic syndrome. For elderly patients, malnutrition and Multiple comorbidities exist, which could affect the discrimination ability of RDW in sepsis. The main purpose of our study was to evaluate the prognostic value of RDW in sepsis among elderly patients.

Methods: This was a retrospective cohort study conducted in emergency department intensive care units (ED- ICU) between April 2015 and November 2015. Elderly patients (>= 65 years old) who were admitted to the ED- ICU with a diagnosis of severe sepsis and/or septic shock were included. The demographic data, biochemistry data, qSOFA, and APACHE II score were compared between survivors and nonsurvivors.

Results: A total of 117 patients was included with mean age 81.5 +- 8.3 years old. The mean APACHE II score was

21.9 +- 7.1. In the multivariate Cox proportional hazards model, RDW level was an independent variable for mor- tality (hazard ratio: 1.18 [1.03-1.35] for each 1% increase in RDW, p = 0.019), after adjusting for CCI, any diag- nosed malignancy, and eGFR. The AUC of RDW in predicting mortality was 0.63 (95% confidence interval [CI]: 0.52-0.74, p = 0.025). In subgroup analysis, for qSOFA b 2, nonsurvivors had higher RDW levels than survivors (17.0 +- 3.3 vs. 15.3 +- 1.4%, p = 0.044).

Conclusions: In our study, RDW was an independent predictor of in-hospital mortality in elderly patients with sepsis. For qSOFA scores b 2, higher RDW levels were associated with poor prognosis. RDW could be a potential parameter used alongside the clinical prediction rules.

(C) 2017

Introduction

Red Cell Distribution Width is calculated as the standard devi- ation of red blood cell (RBC) volume divided by the mean corpuscular volume (MCV) and is a quantitate expression of anisocytosis. RDW is a common hematologic parameter and a part of the standard complete blood count which is measured among hospitalized patients. It has been used as a prognostic biomarker in hypertension [1], coronary dis- ease [2,3], stroke [4], pulmonary hypertension [5], and acute kidney inju- ry [6]. RDW is also related to all-cause mortality in the general population and is even independently associated with Nutritional status [7].

* Corresponding author at: Graduate Institute of Injury Prevention and Control, College of Public Health, Taipei Medical University, Taipei City, Taiwan.

E-mail address: [email protected] (C.-K. Chang).

The exact pathophysiological mechanism underlying RDW and clin- ical outcomes is not well-understood currently. Elevated RDW has been associated with inflammatory biomarkers [8] and RDW increases with increasing oxidative stress [9]. Inflammation impairs iron metabolism, promotes RBC apoptosis, reduces erythropoietin production, and has a myelosuppressive effect [10,11]. In a previous study, it has been report- ed that RDW has moderate discriminative power for mortality in criti- cally-ill patients, including medical and surgical patients [12], which indicates that RDW could be a potential biomarker for assessing the se- verity of sepsis. Increase in RDW from baseline levels during the first three days after ED admission has been associated with mortality in se- vere sepsis or septic shock [13]. In that study, the mean population age was lower than 65 years. For elderly patients, malnutrition and multiple comorbidities exist, such as chronic heart failure and chronic kidney dis- ease which have been related to the increase in RDW [14]. This could af- fect the discrimination ability of RDW in sepsis. The main purpose of our

https://doi.org/10.1016/j.ajem.2017.10.056

0735-6757/(C) 2017

Biochemical parameters“>study was to evaluate the prognostic value of RDW in sepsis among elderly patients.

Materials and methods

Study design

This was a retrospective analysis conducted between April 2015 and November 2015 at the Taipei Medical university hospital, a university- based teaching hospital. The study was conducted in the Emergency Department’s intensive care unit (ED-ICU). This study was approved by the Institutional Review Board of the Taipei Medical University, and the requirement for informed patient consent was waived.

Elderly patients (>= 65 years old) who were admitted to the ED-ICU from the emergency department with a diagnosis of severe sepsis

and/or septic shock according to the International Sepsis Definitions Conference criteria [15] were included in the study. Exclusion criteria included trauma patients, patients transferred from the general wards and postoperative patients.

Demographic data for all the study patients were obtained from the electronic medical records and included age, gender, height, body weight, body mass index (BMI), Charlson Comorbidity Index (CCI) [16] to represent the comorbidities in different illness categories (malignan- cy, metastatic or hematologic malignancies, cardiovascular disease, renal insufficiency, hepatic insufficiency, stroke, respiratory insufficien- cy, and diabetes mellitus), primary site of infection, and Acute Physiolo- gy and Chronic Health Evaluation (APACHE) II score [17] at ICU admission obtained within 24 h.

Laboratory parameters such as white blood cell count, hemo- globin (Hb) level, hematocrit (Hct), platelet count, sodium, potassium, total bilirubin, albumin, c-reactive protein (CRP), creatinine, Estimated glomerular filtration rate and blood gas were measured at initial presentation in the emergency department. RDW values, as a part of the complete blood cell count analysis, were extracted, and compared to the normal laboratory range of RDW in our hospital (11.5% to 14.5%). The main outcome of our study was in-hospital mortality.

Subgroup analysis

The Third International Consensus Definitions for Sepsis and Septic Shock (Sepsis-3) in 2016 introduced the quick Sequential (Sepsis-relat- ed) Organ Failure Assessment (qSOFA) score for sepsis definition, which assesses physiologic parameters, as hypotension (systolic blood pres- sure <= 100 mm Hg), tachypnea (respiratory rate >= 22/min), and altered mental status [18]. A qSOFA score of 2 or more points can be identified as a poor outcome of sepsis. We also compared the qSOFA score (range,

0 [best] to 3 [worst] points) calculated based on the ED’s triage physio- logical parameters with RDW level and mortality.

Statistical analyses

All numeric variables were assessed for normality using the Kolmo- gorov-Smirnov test. Continuous variables were expressed as means and standard deviations, applying the Mann-Whitney U test for compari- sons. Categorical variables, expressed numerically and as proportions, were compared as applicable via chi-square test or Fisher’s exact test. Variables in univariate analysis with a p-value b 0.10 were entered into the multivariate analysis. The multivariate analysis was conducted using Cox proportional hazards model.

A receiver-operating characteristic (ROC) curve was generated for predictive accuracy of RDW level, and the area under the curve (AUC) was calculated. Significance was set at p b 0.05 in two-sided testing. All analyses were performed using the SPSS version 22 (IBM Corp., Armonk, NY).

Results

Baseline characteristics

A total of 117 patients were included in the study. The mean age was

81.5 +- 8.3 years, and 59.8% (70/117) of the patients were male. The mean length of stay (LOS) in ICU was 5.7 +- 5.8 days. In our cohort, 82.9% (97/117) of patients had underlying cardiovascular disease, 23.9% (28/117) had malignancies, 40.2% (47/117) had diabetes, 40.2% (47/117) had cerebrovascular disease, and 21.4% (25/117) had chronic kidney disease. The mean APACHE II score in our cohort was 21.9 +-

7.1. The primary sites of infection were as follows: 62.3% (73/117) were pneumonia, 22.2% (26/117) were intra-Abdominal infections, 14.5% (17/117) were urinary tract infections, and only one was menin- gitis. Non-survivors had shorter LOS in ICU (4.5 +- 4.1 days vs. 6.2 +- 6.4 days, p = 0.045) and total LOS (26.4 +- 39.0 days vs. 11.7 +- 14.1 days, p b 0.001). The patients who survived to discharge had a lower prevalence of underlying malignancy with lower CCI (3.6 +- 2.2 vs. 5.1 +- 3.4, p = 0.056). There were no significant differences in age, gender, BMI, APACHE II score, and primary site of infection. Detailed de- mographic characteristics were shown in Table 1.

Biochemical parameters

Survivors had lower RDW levels (15.5 +- 1.9% vs. 16.7 +- 2.8%, p = 0.025), and higher eGFR (66.56 +- 53.91 ml/min/1.73 m² vs. 41.48 +- 32.20 ml/min/1.73 m², p = 0.007) compared to non-survivors. There was no significant difference in RBC count, Hb, platelet count, CRP, and Albumin level (Table 1).

Other parameters which were incorporated in the APACHE II scoring system are shown in Table 1. There was no significant difference in APACHE II score between the two groups (21.5 +- 5.9 vs. 22.8 +- 9.2, p

= 0.275).

RDW and mortality

In Table 2, higher RDW values, higher CCIs, previous diagnosis of ma- lignancy, and lower eGFR, were associated with mortality. In the multi- variate Cox proportional hazards model, RDW level was an independent variable for mortality (hazard ratio: 1.18 [1.03-1.35] for each 1% in- crease in RDW, p = 0.019), after adjusting for CCI, any diagnosed malig- nancy, and eGFR. The AUC of RDW in predicting mortality was 0.63 (95% confidence interval (CI): = 0.52-0.74, p = 0.025).

Subgroup analysis

Out of 117 patients, 49 (41.9%) had qSOFA scores >= 2. Higher RDW levels in the qSOFA >=2 group was observed, but there was no significant difference between the qSOFA score b 2 and >=2 (15.7 +- 2.2% vs. 16.0 +-

2.4%, p = 0.753) groups. In the group with qSOFA b 2, non-survivors had higher RDW levels than survivors (17.0 +- 3.3% vs. 15.3 +- 1.4%, p = 0.044). For the qSOFA >= 2 group, non-survivors had slightly higher RDW levels, but there was no significant difference (16.3 +- 2.3% vs. 15.8 +- 2.5%, p = 0.254) (Fig. 1). Discrimination of in-hospital mortality using qSOFA was not satisfactory (AUC, 0.56; 95% CI: 0.45-0.67, p =

0.313).

Discussion

In our study, the RDW level could be an independent predictive fac- tor for mortality in elderly patients admitted to the ICU with sepsis. After adjusting for comorbidities and renal function, for each 1% in- crease in RDW level as a continuous variable in the multivariate Cox proportional hazards model, mortality rate increased by 18%. The ROC curve showed that RDW had moderate discriminative power for in-hos- pital mortality.

Table 1

Demographic data of the study population.

Variables	Total (n = 117)	Survivors (n = 81)	Non-survivors (n = 36)	p-Value
Demographic parameters
Gender (male) – no. (%)	70 (59.8%)	47 (58.0%)	23 (63.9%)	0.550
Age – yr (mean +- S.D.)	81.5 +- 8.3	80.9 +- 7.3	82.7 +- 10.3	0.281
Body weight – kg (mean +- S.D.)	57.2 +- 12.5	58.3 +- 13.3	54.8 +- 10.4	0.293
Body height – cm (mean +- S.D.)	160.5 +- 8.8	161.4 +- 8.8	158.6 +- 8.6	0.152
BMI – kg/m2 (mean +- S.D.)	22.25 +- 4.3	22.3 +- 4.5	21.8 +- 4.0	0.666
Primary site of infections
Pneumonia – no. (%)	73 (62.4%)	49 (60.5%)	24 (66.7%)	0.525
Urinary tract infections – no. (%)	17 (14.5%)	14 (17.3%)	3 (8.3%)	0.205
intra-abdominal infections – no. (%)	26 (22.2%)	17 (21.0%)	9 (25.0%)	0.630
Other – no. (%)	1 (0.9%)	0	1 (2.8%)	0.308a
qSOFA score				0.235
b2 – no. (%)	68 (58.1%)	50 (61.7%)	18 (50.0%)
? 2 – no. (%)	49 (41.9%)	31 (38.3%)	18 (50.0%)
Comorbidities
Malignancies – no. (%)	28 (23.9%)	12 (14.8%)	16 (44.4%)	0.001
Diabetes mellitus – no. (%)	47 (40.2%)	33 (40.7%)	14 (38.9%)	0.850
Cerebrovascular disease – no. (%)	47 (40.2%)	34 (42.0%)	13 (36.1%)	0.550
Cardiovascular disease – no. (%)	97 (82.9%)	68 (84.0%)	29 (80.6%)	0.259
Hepatic insufficiency – no. (%)	6 (5.1%)	6 (7.4%)	0	0.175a
Renal insufficiency or dialysis – no. (%)	25 (21.4%)	15 (18.5%)	10 (27.8%)	0.653
Respiratory insufficiency – no. (%)	43 (6.8%)	32 (39.5%)	11 (30.6%)	0.354
Charlson comorbidity index – (mean +- S.D.)	4.1 +- 2.7	3.6 +- 2.2	5.1 +- 3.4	0.056
Biochemistry parameters
Laboratory parameters RBC – x103/uL (mean +- S.D.)	3.8 +- 0.7	3.9 +- 0.7	3.7 +- 0.8	0.186
Hb – g/dL (mean +- S.D.)	11.5 +- 2.1	11.6 +- 2.0	11.2 +- 2.3	0.408
RDW – % (mean +- S.D.)	15.9 +- 2.3	15.5 +- 1.9	16.7 +- 2.8	0.025
Platelet – x103/uL (mean +- S.D.)	215.3 +- 100.5	217.2 +- 99.7	211.0 +- 103.6	0.593
CRP – mg/dL (mean +- S.D.)	11.7 +- 10.6	11.6 +- 10.4	12.0 +- 11.3	0.765
Albumin – g/dl (mean +- S.D.)	3.1 +- 0.5	3.1 +- 0.5	3.2 +- 0.5	0.261
eGFR – mL/min/1.73m2 (mean +- S.D.)	58.7 +- 49.4	66.6 +- 53.9	41.5 +- 32.2	0.007
Laboratory parameters included in APACHE II score WBC – x103/uL (mean +- S.D.)	13.9 +- 7.2	15.1 +- 7.6	11.2 +- 5.6	0.003
Hct – % (mean +- S.D.)	34.5 +- 6.1	37.9 +- 5.7	33.6 +- 7.0	0.379
pH – (mean +- S.D.)	7.38 +- 0.90	7.39 +- 0.09	7.35 +- 0.09	0.015
pO2 – mm Hg (mean +- S.D.)	119.7 +- 90.2	125.6 +- 101.7	106.5 +- 55.7	0.887
Creatinine – mg/dl (mean +- S.D.)	2.8 +- 8.2	3.0 +- 9.8	2.5 +- 1.9	0.017
Na – mEq/L (mean +- S.D.)	135.6 +- 10.0	134.7 +- 8.5	137.7 +- 12.7	0.056
K – mEq/L (mean +- S.D.)	4.3 +- 1.0	4.3 +- 1.1	4.5 +- 0.9	0.107
APACHEII score – (mean +- S.D.)	21.9 +- 7.1	21.5 +- 5.9	22.8 +- 9.2	0.275
LOS in ICU – days (mean +- S.D.)	5.7 +- 5.8	6.2 +- 6.4	4.5 +- 4.1	0.045
Total LOS – days (mean +- S.D.)	21.9 +- 34.0	26.4 +- 39.0	11.7 +- 14.1	0.000

^a Fisher’s exact test Data are expressed as mean +- standard deviation or number (percentage) qSOFA, quick Sequential Organ Failure Assessment; BMI, body mass index; RBC, red blood cell; WBC, white blood cell; Hb, hemoglobin; Hct, hematocrit; CRP, C-reactive protein; eGFR, estimated glomerular filtration rate; LOS, length of stay; ICU, intensive care unit.

RDW is a parameter that measures the range of variation of red blood cell size, which is a common and inexpensive biomarker in critical illness, as a part of the complete blood count analysis. Higher RDW levels indicate greater variation in size, which is called anisocytosis; it could be used for the differential diagnosis of nutritional deficiency-re- lated anemia due to iron [19], folic acid, and vitamin B12 deficiency [20]. RDW has been associated with the severity and/or mortality in many diseases, including cardiovascular diseases [21,22], chronic obstructive pulmonary disease [23], pulmonary hypertension [5], rheumatoid ar- thritis [24], dementia [25], and malignancy [26].

Previous studies have reported that higher RDW was also associated with poorer outcome in community-acquired pneumonia [27], gram-

negative bacteremia [28], severe sepsis and septic shock [29], and all- cause mortality in ICU patients [12]. There is also an association be- tween RDW and pro-inflammatory cytokines [30], tumor necrosis factor alpha [31], oxidative stress as Reactive oxygen species, and superoxide dismutase [32]. The inflammatory response alters the half-life of red blood cells, erythropoiesis, disorders of iron metabolism, and increases hemolysis, which results in impaired hematopoiesis and increases RBC size heterogeneity. Decreased levels of anti-oxidants are also associated with increasing RDW values [33]. Persistently higher RDW is also found in non-surviving patients with sepsis during the first week [31].

A previous study performed in geriatric wards demonstrated that higher RDW was associated with higher all-cause mortality risk during

Table 2

Univariate and multivariate Cox proportional hazard model.

	Univariate				Multivariate
	Hazard ratio	95% confidence interval	p-Value		Hazard ratio	95% confidence interval	p-Value
RDW – per 1%	1.18	1.05-1.33	0.006		1.18	1.03-1.35	0.019
Charlson comorbidity index – per 1 point	1.12	1.01-1.24	0.03		0.98	0.87-1.10	0.713
Malignancy	3.02	1.55-5.85	0.001		4.32	2.05-9.10	b0.001
eGFR	0.99	0.98-0.99	0.023		0.99	0.98-0.99	0.004
APACHE II score – per 1 point	1.02	0.97-1.07	0.492		–	–	–
qSOFA ? 2	1.48	0.77-2.85	0.240		–	–	–

Fig. 1. Subgroup analysis of qSOFA score with RDW level and mortality.

a 5-year follow-up [34]. In the Geriatric population, there are more com- plex factors that could affect the heterogeneity of RBC size, such as age, liver or renal dysfunction, malnutrition, cancer, thyroid disease, and acute or chronic inflammatory response [35,36]. Higher RDW levels in the geriatric population are associated with higher mortality and could be a consequence of higher levels of inflammatory cytokines and/or lesser overall health status for dealing with infection and tissue repair.

The APACHE II [17], a physiology-based scoring system, comprises 12 routine variables including blood gas, vital signs, electrolytes, creati- nine, hematocrit and WBC count. The APACHE II score is commonly ap- plied within 24 h after ICU admission for Mortality prediction in critically-ill patients. In our population, there was no significant differ- ence in APACHE II score between survivors and non-survivors. Also, the APACHE II scoring system includes creatinine instead of eGFR, which is calculated from serum Creatinine levels taking into consider- ation age, gender and weight. In the elderly, who might be malnour- ished, lower muscle mass and have lower creatinine production, serum creatinine is an insensitive indicator of renal function [37]. A higher prevalence of lower eGFR is inversely related with RDW level in- dependently of age, gender [38]. After adjusting for a wide variety of co- variates including comorbidity and laboratory parameters, RDW was still a predictor of mortality, which is important in treating Geriatric patients.

The qSOFA score was introduced in the Sepsis-3 criteria [18] to assist clinicians in identifying patients with suspected infection with poorer outcomes. The qSOFA score comprises only three clinical elements: re- spiratory rate >= 22/min, altered mentation, systolic blood pressure

<= 100 mm Hg, meaning that it is simple and easy to apply. Prompt

early identification and management of sepsis prevents further deterio- ration and reduces morbidity and mortality. An increase in qSOFA score of 2 or more indicated before ICU admission has good performance in predicting mortality and ICU-free days [39]. In the qSOFA, altered men- tation is defined by either a Glasgow Coma Scale score of b 15 or

<= 13 [18]. Delirium is common in elderly patients and is often precipitat-

ed by infection [40]. In our study, 85.5% of patients had GCS scores b 15; moreover, older adults have more complex and less physiologic reserve. For a normotensive patient with delirium who presented to the ED with infection, the qSOFA score was only 1. If the patient deteriorated and re- quired intensive care, qSOFA criteria could be less sensitive and not sat- isfactory for triage decision. In our results, for the qSOFA b 2 group, RDW could help discriminate the in-hospital mortality. RDW as a commonly

evaluated laboratory parameter could help physicians discriminate prognosis. Higher RDW level could predict poor outcome.

Our study included a mostly aging population with a mean age of 81 years. There was no significant difference in age, underlying cardio- vascular, cerebrovascular disease, and respiratory or hepatic insufficien- cy between survivors and non-survivors. The prognostic value of RDW is independent of chronic health conditions. RDW could be a potential parameter used along with the clinical prediction rules.

Study limitations

This was a single-center study with limited case numbers. Also, this was a retrospective study, and all the limitations of a retrospective re- view could be inherent in our study. Missing data, incomplete docu- mentation, recall bias, and information bias could also be encountered and possibly affect the results. Selection bias may exist in our study. We included patients admitted to the ED-ICU; whether the results ap- plied to internal medicine-ICU, surgical-ICU, cardiac-ICU, or other spe- cialty ICUs had yet to be determined. Our physicians were all emergency physicians who were trained in critical care so that the treat- ment policy would be similar to that of other ICUs. An extension of the study period to include more patients or a further prospective study could increase the amount of data available for analysis. RDW could be influenced by folate, iron, and vitamin B12. In most facilities, these var- iables were not routinely evaluated. Furthermore, receiving a blood transfusion before blood sampling could affect the RDW. Hence, we choose the laboratory values at initial presentation at the emergency department instead of at ICU admission to avoid the possibility of this confounding issue, and there were a few cases which were transferred from other hospitals. Moreover, due to the retrospective nature of our study, patients underwent blood sampling at different time intervals depending on their conditions after admitted to ICU. It was difficult to compare with the variations in RDW. However, since the purpose of our study was to enable physicians to recognize sepsis early, RDW was estimated at initial presentation to the ED. For elderly patients with low initial qSOFA scores, we can add the RDW level to predict the outcome.

Conclusions

RDW is a commonly reported laboratory parameter in clinical set- tings. In our study, RDW was an independent predictor of in-hospital mortality in elderly patients with sepsis. For qSOFA scores b 2, higher RDW levels were associated with poor prognosis. RDW could be a po- tential parameter used alongside the clinical prediction rules, which could help clinicians have more confidence in recognizing and predicting the outcome at the initial presentation of sepsis in the geriat- ric population.

Acknowledgments

None.

Conflict of interest statement

The authors declare no conflicts of interest or sources of funding for this study.

References

1. Bilal A, Farooq JH, Kiani I, Assad S, Ghazanfar H, Ahmed I. Importance of mean red cell distribution width in Hypertensive patients. Cureus 2016;8:e902.
2. Tonelli M, Sacks F, Arnold M, Moye L, Davis B, Pfeffer M. For the cholesterol and re- current events (CARE) trial investigators. Relation between red blood cell distribu- tion width and cardiovascular event rate in people with coronary disease. Circulation 2008;117:163-8.
3. Sangoi MB, Da Silva SH, da Silva JE, et al. Relation between red blood cell distribution width and mortality after acute myocardial infarction. Int J Cardiol 2011;146: 278-80.
4. Ani C, Ovbiagele B. Elevated red blood cell distribution width predicts mortality in persons with known stroke. J Neurol Sci 2009;277:103-8.
5. Hampole CV, Mehrotra AK, Thenappan T, Gomberg-Maitland M, Shah SJ. Usefulness of red cell distribution width as a prognostic marker in pulmonary hypertension. Am J Cardiol 2009;104:868-72.
6. HJ Oh, Park JT, Kim JK, Yoo DE, Kim SJ, Han SH, et al. Red blood cell distribution width is an independent predictor of mortality in acute kidney injury patients treated with Continuous renal replacement therapy. Nephrol Dial Transplant 2012;27:589-94.
7. Patel KV, Ferrucci L, Ershler WB, Longo DL, Guralnik JM. Red cell distribution width and the risk of death in middle-aged and older adults. Arch Intern Med 2009;169: 515-23.
8. Lippi G, Targher G, Montagnana M, Salvagno GL, Zoppini G, Guidi GC. Relation be- tween red blood cell distribution width and inflammatory biomarkers in a large co- hort of unselected outpatients. Arch Pathol Lab Med 2009;133:628-32.
9. Friedman JS, Lopez MF, Fleming MD, Rivera A, Martin FM, Welsh ML, et al. SOD2-de- ficiency anemia: protein oxidation and altered protein expression reveal targets of damage, stress response, and antioxidant responsiveness. Blood 2004;104:2565-73.
10. Ghali JK. Anemia and heart failure. Curr Opin Cardiol 2009;24:172-8.
11. Weiss G, Goodnough LT. Anemia of chronic disease. N Engl J Med 2005;352: 1011-23.
12. Bazick HS, Chang D, Mahadevappa K, Gibbons FK, Christopher KB. Red cell distribu- tion width and all-cause mortality in critically ill patients. Crit Care Med 2011;39: 1913-21.
13. Kim CH, Park JT, Kim EJ, Han JH, Han JS, Choi JY, et al. An increase in red blood cell distribution width from baseline predicts mortality in patients with severe sepsis or septic shock. Crit Care 2013;17:R282.
14. van Breda F, Emans ME, van der Putten K, Braam B, van Ittersum FJ, Kraaijenhagen RJ, et al. Relation between red cell distribution width and fibroblast growth factor 23 cleaving in patients with chronic kidney disease and heart failure. PLoS One 2015;10:e0128994.
15. Dellinger RP, Levy MM, Rhodes A, Annane D, Gerlach H, Opal SM, et al. Surviving sepsis campaign: international guidelines for management of severe sepsis and sep- tic shock: 2012. Crit Care Med 2013;41:580-637.
16. Charlson ME, Pompei P, Ales KL, MacKenzie CR. A new method of classifying prog- nostic comorbidity in longitudinal studies: development and validation. J Chronic Dis 1987;40:373-83.
17. Knaus WA, Draper EA, Wagner DP, Zimmerman JE. APACHE II: a severity of disease classification system. Crit Care Med 1985;13:818-29.
18. Singer M, Deutschman CS, Seymour CW, Shankar-Hari M, Annane D, Bauer M, et al. The third international consensus definitions for sepsis and septic shock (Sepsis-3). JAMA 2016;315:801-10.
19. Viswanath D, Hegde R, Murthy V, Nagashree S, Shah R. Red cell distribution width in the diagnosis of Iron deficiency anemia. Indian J Pediatr 2001;68:1117-9.
20. Evans TC, Jehle D. The red blood cell distribution width. J Emerg Med 1991;9(Suppl. 1):71-4.
21. Anderson JL, Ronnow BS, Horne BD, Carlquist JF, May HT, Bair TL, et al. Usefulness of a complete blood count-derived risk score to predict incident mortality in patients with suspected cardiovascular disease. Am J Cardiol 2007;99:169-74.
22. Sangoi MB, Guarda Ndos S, Rodel AP, Zorzo P, Borges PO, Cargnin LP, et al. Prognostic value of red blood cell distribution width in prediction of in-hospital mortality in pa- tients with acute myocardial infarction. Clin Lab 2014;60:1351-6.
23. Seyhan EC, Ozgul MA, Tutar N, Omur I, Uysal A, Altin S. Red blood cell distribution and survival in patients with chronic obstructive pulmonary disease. COPD 2013; 10:416-24.
24. Yunchun L, Yue W, Jun FZ, Qizhu S, Liumei D. Clinical significance of red blood cell distribution width and inflammatory factors for the disease activity in rheumatoid arthritis. Clin Lab 2016;62:2327-31.
25. Weuve J, Mendes de Leon CF, Bennett DA, Dong X, Evans DA. The red cell distribu- tion width and anemia in association with prevalent dementia. Alzheimer Dis Assoc Disord 2014;28:99-105.
26. Montagnana M, Danese E. Red cell distribution width and cancer. Ann Transl Med 2016;4:399.
27. Lee JH, Chung HJ, Kim K, Jo YH, Rhee JE, Kim YJ, et al. Red cell distribution width as a prognostic marker in patients with community-acquired pneumonia. Am J Emerg Med 2013;31:72-9.
28. NS Ku, Kim HW, HJ Oh, Kim YC, Kim MH, Song JE, et al. Red blood cell distribution width is an independent predictor of mortality in patients with gram-negative bac- teremia. Shock 2012;38:123-7.
29. Jo YH, Kim K, Lee JH, Kang C, Kim T, Park HM, et al. Red cell distribution width is a prognostic factor in severe sepsis and septic shock. Am J Emerg Med 2013;31:545-8.
30. Pierce CN, Larson DF. Inflammatory cytokine inhibition of erythropoiesis in patients implanted with a mechanical circulatory assist device. Perfusion 2005;20:83-90.
31. Lorente L, Martin MM, Abreu-Gonzalez P, Sole-Violan J, Ferreres J, Labarta L, et al. Red blood cell distribution width during the first week is associated with severity and mortality in septic patients. PLoS One 2014;9:e105436.
32. Ghaffari S. Oxidative stress in the regulation of normal and neoplastic hematopoie- sis. Antioxid Redox Signal 2008;10:1923-40.
33. Semba RD, Patel KV, Ferrucci L, Sun K, Roy CN, Guralnik JM, et al. Serum antioxidants and inflammation predict red cell distribution width in older women: the Women’s Health and Aging Study I. Clin Nutr 2010;29:600-4.
34. Martinez-Velilla N, Ibanez B, Cambra K, Alonso-Renedo J. Red blood cell distribution width, multimorbidity, and the risk of death in hospitalized older patients. Age (Dordr) 2012;34:717-23.
35. Hoffmann JJ, Nabbe KC, van den Broek NM. Effect of age and gender on reference in- tervals of Red blood cell distribution width and mean red cell volume (MCV). Clin Chem Lab Med 2015;53:2015-9.
36. Luo R, Hu J, Jiang L, Zhang M. Prognostic value of red blood cell distribution width in non-cardiovascular critically or acutely patients: a systematic review. PLoS One 2016;11:e0167000.
37. Lubran MM. Renal function in the elderly. Ann Clin Lab Sci 1995;25:122-33.
38. Lippi G, Targher G, Montagnana M, Salvagno GL, Zoppini G, Guidi GC. Relationship between red blood cell distribution width and kidney function tests in a large cohort of unselected outpatients. Scand J Clin Lab Invest 2008;68:745-8.
39. Finkelsztein EJ, Jones DS, Ma KC, Pabon MA, Delgado T, Nakahira K, et al. Comparison of qSOFA and SIRS for predicting adverse outcomes of patients with suspicion of sep- sis outside the intensive care unit. Crit Care 2017;21:73.
40. O’Brien ME. Dealing with delirium. A common problem in hospitalized elderly pa- tients. Postgrad Med 1992;91:463-70.