Red blood cell distribution width and neutrophil to lymphocyte ratio are associated with outcomes of adult subarachnoid haemorrhage patients admitted to intensive care unit

Abstract

Background

Red blood cell distribution width (RDW) and neutrophil to lymphocyte ratio (NLR) have been reported to be associated with outcomes of acute cerebral infarction. However, their prognostic value in patients with subarachnoid haemorrhage (SAH) remains largely unknown. The aim of this study was to investigate the prognostic value of RDW and NLR in SAH patients.

Methods

Medical records of adult SAH patients admitted to intensive care unit (ICU) were extracted from Multiparameter Intelligent Monitoring in Intensive Care II (MIMIC II, version 2.6), a publicly accessible ICU database. Prognostic value of RDW and NLR was analysed using logistic regression model, Kaplan-Meier curve analysis and Cox regression model.

Results

A total of 274 SAH patients were included. Patients died in hospital had significantly higher RDW and NLR. RDW and NLR were significantly associated with hospital death, with adjusted odds ratios of 1.39 (95% CI, 1.06–1.82) and 1.04 (95% CI, 1.00–1.08), respectively. Furthermore, increased RDW and NLR were associated with higher one-year mortality, with an adjusted hazard ratio of 1.20 (95% CI, 1.02–1.41) for per 1% increased RDW and 1.03 (95% CI, 1.00–1.05) for per 1 increased NLR.

Conclusion

RDW and NLR are useful indices to evaluate the outcomes of ICU admitted patients with SAH.

Keywords

Red blood cell distribution width neutrophil to lymphocyte ratio subarachnoid haemorrhage hospital death one-year mortality

Introduction

Red blood cell distribution width (RDW) is a routinely tested index to describe the size variation of red blood cells.¹ Over the past decades, RDW has been proved to be a useful index in exploring the aetiology of anaemia, especially in differentiating thalassaemia and iron deficiency-related anaemia.² However, recent studies revealed that RDW is useful in estimating the risk, prognosis or disease activity of non-haematologic diseases, such as cardiovascular diseases,^3–7 liver diseases,^8,9 autoimmune diseases^10–12 or stroke.¹³ In patients with acute subarachnoid haemorrhage (SAH)^14,15 or acute cerebral infarction,¹⁶ RDW was proved to be associated with short-term and functional outcome of patients. However, whether RDW is associated with long-term outcome of SAH remains unknown.

Neutrophil to lymphocyte ratio (NLR), which is defined as the ratio of neutrophils percentage to lymphocytes percentage, is an easily acquired haematological index. Similar to RDW, NLR has been revealed to be a prognostic factor for patients with acute cerebral infarction.¹⁷ However, whether NLR can be used in evaluating the prognosis of SAH remains largely unknown.

In this study, with the use of a large clinical database named Multiparameter Intelligent Monitoring in Intensive Care II (MIMIC II, version 2.6), we investigated the prognostic value of RDW and NLR in adult patients with SAH.

Materials and methods

Summary of database

MIMIC II is a large clinical database consists of more than 30,000 patients admitted to intensive care unit (ICU).¹⁸ All these patients were enrolled in Beth Israel Deaconess Medical Center (Boston, MA) from 2001 to 2008. This database is composed of 38 tables, which record patients’ information, such as demographics characteristics, treatment outcomes and laboratory tests. All patients in this database are de-identified to protect patients’ privacy.

After completing a training course on the website of National Institutes of Health (NIH), an author (Z-D Hu, certification number: 1678079) was approved to access the database. The accessing process was well documented by Zhang.¹⁹

Informed consent was waived since the establishment of this database is approved by the Institutional Review Boards of the Massachusetts Institute of Technology (Cambridge, MA) and Beth Israel Deaconess Medical Center (Boston, MA).

Data extraction

Similar to our previous studies,^7,20,21 data were extracted using structure query language (SQL) with pgAdmin (version 1.12.3), an open source development and administration platform for PostgreSQL. Five tables in MIMIC II database were used in our study, including ICD9, COMORBIDITY_SCORES, LABEVENTS, D_LABITEMS and ICUSTAY_DETAIL. ICD9 was used to define the patients with SAH (code = 430 and sequence = 1). D_LABITEMS and LABEVENTS were used to extract results of laboratory tests. ICUSTAY_DETAIL was used to extract age group, clinical outcomes, simplified acute physiology score (SAPS) I.²² In MIMIC II database, adult is defined as patients more than 15 years old. As to patients received a laboratory test more than one time during their hospitalization, only the initial test results were analysed. Patients who have been previously admitted to ICU were excluded.

Statistical analysis

Continuous variables were presented as mean and standard deviation (SD) and compared by Mann-Whitney U test. Categorical variables were compared using Chi-square test. Logistic regression model, Cox regression model and Kaplan-Meier survival curve analysis were used to evaluate the prognostic value of RDW, SAPS, NLR and platelet. All statistical analyses were performed using SPSS 17.0. P value less than 0.05 was defined as statistically significant.

Results

Characteristics of subjects

A total of 274 subjects were included in this study, and their clinical characteristics are listed in Table 1. The subjects died in hospital (n = 61) were older, with significantly higher white blood cell (WBC), NLR, RDW, creatinine and SAPS score. Their lymphocyte, platelet and albumin were significantly lower.

Table 1.

Characteristics of subjects.

	All (n = 274)		Hospital death
	Results	n	No (n = 213)	n	Yes (n = 61)	n	P
Age (years)	59 ± 16	274	57 ± 14	213	68 ± 17	61	<0.001
Gender (M/F)	110/164	274	84/129	213	26/35	61	0.65
WBC (10⁹/L)	12.6 ± 5.6	271	12.2 ± 5.4	212	14.1 ± 6.1	59	0.02
Lymphocyte (%)	15.8 ± 10.9	214	16.5 ± 10.6	169	12.8 ± 11.4	45	<0.001
Neutrophil (%)	78.0 ± 12.5	214	77.5 ± 12.2	169	80.1 ± 13.5	45	0.06
NLR	8.7 ± 9.2	213	7.8 ± 8.7	168	11.8 ± 10.7	45	<0.001
Monocyte (%)	3.9 ± 2.0	214	3.83 ± 1.90	169	4.33 ± 2.33	45	0.29
RBC (10¹²/L)	4.26 ± 0.56	271	4.28 ± 0.54	212	4.17 ± 0.63	59	0.27
MCV (fl)	89.3 ± 6.1	271	89.3 ± 6.0	212	89.3 ± 6.3	59	0.56
MCH (pg)	30.7 ± 2.4	271	30.8 ± 2.3	212	30.3 ± 2.8	59	0.39
RDW (%)	13.50 ± 1.36	271	13.4 ± 1.1	212	14.0 ± 1.9	59	0.02
Platelet (10⁹/L)	254 ± 95	271	259 ± 89	212	235 ± 111	59	0.02
Creatinine (μmol/L)	76 ± 44	271	72 ± 23	212	92 ± 83	59	0.01
Albumin (g/L)	35.5 ± 5.9	162	36.3 ± 5.5	128	32.7 ± 6.4	34	<0.01
SAPS score	12.5 ± 5.6	261	10.9 ± 4.7	207	18.7 ± 4.4	54	<0.01
Heart failure (Y/N)	26/246	272	21/190	211	5/56	61	0.68
Hypertension (Y/N)	128/144	272	99/112	211	29/32	61	0.93
Diabetes (Y/N)	32/240	272	20/191	211	12/49	61	0.03
Anaemia (Y/N)	16/256	272	14/197	211	2/59	61	0.33

Note: Values are presented as mean and standard deviation or number of subjects, where appropriate.

WBC: white blood cell; NLR: neutrophil to lymphocyte ratio; RBC: red blood cell; MCV: mean corpuscular volume; MCH: mean corpuscular haemoglobin; RDW: red blood cell distribution width; SAPS: simplified acute physiology score.

Prognostic factors associated with hospital death

Next, we analysed the factors associated with hospital death among the subjects. We only analysed the prognostic value of RDW, NLR, SAPS and platelet in a logistic regression model. Age, WBC, renal function, comorbidities were excluded because they are included in SAPS score. The interactions between these three parameters and SAPS score may bias the estimates of the model.²³ Albumin was excluded because the results were only available in 162 subjects.

Table 2 lists the effects of RDW, NLR, SAPS and platelet on hospital death. In a univariable logistic regression model, RDW, NLR and SAPS were found significantly associated with hospital death. Therefore, these factors were included in a multivariable logistic regression model, and all of them were found significantly associated with hospital death.

Table 2.

Logistic regression analysis for hospital death.

	Univariable		Multivariable
	OR (95% CI)	P	OR (95% CI)	P
RDW per 1%	1.35 (1.11–1.64)	<0.01	1.39 (1.06–1.82)	0.02
NLR per 1	1.04 (1.00–1.08)	0.03	1.04 (1.00–1.08)	0.04
SAPS per 1	1.41 (1.28–1.55)	<0.01	1.42 (1.26–1.59)	<0.01
Platelet per 50 × 10⁹/L	0.85 (0.71–1.02)	0.08	–	–

NLR: neutrophil to lymphocyte ratio; RDW: red blood cell distribution width; SAPS: simplified acute physiology score; OR: odds ratio.

Effects of RDW, NLR and SAPS on one-year mortality

Figure 1 is a Kaplan-Meier survival curve for the subjects. Figure 1(a) shows that subjects with RDW higher than 15% (upper limit of reference interval²⁴) had significantly poorer outcomes. We also found that subjects with NLR higher than 6.86 (median of NLR across all subjects) had significantly poorer outcomes, as shown in Figure 1(b).

Figure 1.

Kaplan-Meier curve analysing prognostic value of RDW (a) and NLR (b).

Prognostic value of RDW, NLR and SAPS were listed in Table 3. We only included these three factors in a Cox regression model because they were independent factors for hospital death, as indicated in Table 2. In both univariable and multivariable Cox regression model, RDW, NLR and SAPS were significantly associated with one-year mortality.

Table 3.

Cox regression model analysis for one-year mortality.

	Univariable		Multivariable
	HR (95% CI)	P	HR (95% CI)	P
RDW per 1%	1.26 (1.11–1.44)	<0.01	1.20 (1.02–1.41)	0.03
NLR per 1	1.02 (1.00–1.04)	0.02	1.03 (1.00–1.05)	0.04
SAPS per 1	1.28 (1.21–1.35)	<0.01	1.30 (1.21–1.40)	<0.01

NLR: neutrophil to lymphocyte ratio; RDW: red blood cell distribution width; SAPS: simplified acute physiology score; HR: hazards ratio.

Discussion

In this study, we found that both RDW and NLR were useful indices to evaluate hospital mortality and one-year mortality in adult SAH patients admitted to ICU, and their prognostic value were independent of SAPS, a well-known score system in evaluating the outcomes of ICU patients.

Mechanisms underlying RDW, NLR and outcomes of SAH patients remain largely unknown. We hypothesized that the prognostic value of NLR and RDW are related to inflammatory response. It is well known that inflammatory response is critically involved in the pathogenesis of SAH.²⁵ Previous studies have found that increased inflammatory markers, such as C-reactive protein (CRP)²⁶ and erythrocyte sedimentation rate (ESR),²⁷ were associated with poor outcomes of SAH. A recent study also revealed that increased NLR is associated with poor outcomes at three months.²⁸ Meanwhile, RDW and NLR have been proved as inflammatory markers. For example, Lippi et al. reported that RDW is positively correlated with inflammatory markers (CRP and ESR) in a cohort of unselected outpatients.²⁹ In healthy individuals, a positive correlation between RDW and inflammatory markers was also observed.³⁰ In our previous study, we found that RDW was correlated with inflammatory markers in primary Sjogren's syndrome¹⁰ and systemic lupus erythematous.¹¹ For NLR, some studies also indicated that it is an inflammatory marker.^31,32 For example, NLR is positively correlated with CRP in autoimmune disease^33,34 and chronic obstructive pulmonary disease (COPD).³⁵ Taken together, these results indicate that inflammation at least partially mediates the association between the outcomes of SAH and both RDW and NLR.

Besides inflammation response, another possible mechanism responsible for increased RDW is hypoxaemia. It is true that hypoxaemia is very common in SAH patients,³⁶ and a recent study found that hypoxaemia can increase RDW.³⁷

Prognostic evaluation is a critical step in managing SAH patients. Previous studies found that SAPS, a well-known tool in evaluating the prognosis of critical ill patients, is useful in evaluating the prognosis of SAH.³⁸ Since the cohort in our study was adult SAH patients admitted in ICU, it is reasonable to evaluate the prognosis of subjects using SAPS. We found that the prognostic value of NLR and RDW was independent of SAPS, indicating that RDW and NLR may provide additional value in evaluating the prognosis of SAH. Actually, RDW and NLR were prognostic factors for critical ill patients.^39–43 Given that RDW and NLR were independent prognostic factors, we suggest clinicians take them into consideration when evaluating the outcomes of SAH.

To date, many novel prognostic markers for SAH have been identified, such as thrombospondin-1⁴⁴ and soluble CD40.⁴⁵ Compared with these markers, NLR and RDW are easily obtained and require no additional costs. In addition, compared with other inflammatory markers which are useful in evaluating the outcomes of SAH (e.g. CRP²⁶), RDW is less affected by environmental factors, such as infection.¹⁰ Because RDW is only determined by the size variation of red blood cell, and the life span of red blood cell is approximately 130 days⁴⁶; therefore, RDW may reflect a long-term inflammatory status of an individual. By contrast, acute phase protein (e.g. CRP) may only reflect short-term inflammatory status of an individual, and thus has higher biological variation. Taken together, it seems that RDW is more reliable than acute phase protein in estimating the inflammatory status of SAH, and therefore may be preferred for estimating the outcomes of SAH.

Our study has two limitations. First, some clinical data, such as CRP, D-Dimer and blood pressure variability during admission, have been proved to be associated with the prognosis of SAH.^47–50 However, these parameters were absent in more than half of subjects in MIMIC II database. Therefore, we cannot analyse whether the prognostic value of NLR and RDW is independent of these factors. Second, this study was performed in SAH subjects admitted to ICU; therefore, the current findings cannot be generalized non-ICU SAH patients.

In conclusion, our study indicated that both RDW and NLR were useful prognostic factors for SAH. Further studies with prospective design and more clinical details are needed to better evaluate the prognostic value of RDW and NLR in SAH patients. Besides, it is also interesting to validate our findings in non-ICU SAH patients.

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

This work was supported by grants from the National Natural Science Foundation of China (No. 81302541) and Key Program of Nanjing Military Command of PLA (No. 15DX007).

Ethical approval

Not applicable.

Guarantor

Z-D H.

Contributorship

Z-DH designed research and extracted data from database; ZJH and YLH performed statistics, wrote the first draft of manuscript. All authors reviewed and edited the manuscript and approved the final version of the manuscript.

References

Salvagno

Sanchis-Gomar

Picanza

et al.

Red blood cell distribution width: a simple parameter with multiple clinical applications. Crit Rev Clin Lab Sci 2015; 52: 86–105.

Leung

Lau

et al.

Evaluation of the new red cell parameters on Beckman Coulter DxH800 in distinguishing iron deficiency anaemia from thalassaemia trait. Int J Lab Hematol 2015; 37: 199–207.

Turcato

Serafini

Dilda

et al.

Red blood cell distribution width independently predicts medium-term mortality and major adverse cardiac events after an acute coronary syndrome. Ann Transl Med 2016; 4: 254–254.

Danese

Lippi

Montagnana

. Red blood cell distribution width and cardiovascular diseases. J Thorac Dis 2015; 7: E402–E411.

Huang

Liu

et al.

Prognostic value of red blood cell distribution width for patients with heart failure: a systematic review and meta-analysis of cohort studies. PLoS One 2014; 9: e104861–e104861.

Liao

Song

et al.

The role of red blood cell distribution width in mortality and cardiovascular risk among patients with coronary artery diseases: a systematic review and meta-analysis. J Thorac Dis 2014; 6: 1429–1440.

Huang

. Lower mean corpuscular hemoglobin concentration is associated with poorer outcomes in intensive care unit admitted patients with acute myocardial infarction. Ann Transl Med 2016; 4: 190–190.

Sun

Wang

et al.

Red blood cell distribution width is a potential prognostic index for liver disease. Clin Chem Lab Med 2013; 51: 1403–1408.

Qiu

et al.

Red blood cell distribution width levels correlate with liver fibrosis and inflammation: a noninvasive serum marker panel to predict the severity of fibrosis and inflammation in patients with hepatitis B. Medicine (Baltimore) 2015; 94: e612–e612.

10.

Sun

Guo

et al.

Red blood cell distribution width and neutrophil/lymphocyte ratio are positively correlated with disease activity in primary Sjogren's syndrome. Clin Biochem 2014; 47: 287–290.

11.

Chen

Zhang

et al.

Red blood cell distribution width is a potential index to assess the disease activity of systemic lupus erythematosus. Clin Chim Acta 2013; 425: 202–205.

12.

. Red blood cell distribution width: a promising index for estimating activity of autoimmune disease. J Lab Precis Med 2016; 1: 5–5.

13.

Lappegard

Ellingsen

Skjelbakken

et al.

Red cell distribution width is associated with future risk of incident stroke. The Tromso Study. Thromb Haemost 2015; 115: 126–134.

14.

Chugh

Nyirjesy

Nawalinski

et al.

Red blood cell distribution width is associated with poor clinical outcome after subarachnoid hemorrhage: a pilot study. Neurocrit Care 2015; 23: 217–224.

15.

Siegler

Marcaccio

Nawalinski

et al.

Elevated red cell distribution width is associated with cerebral infarction in aneurysmal subarachnoid hemorrhage. Neurocrit Care. Epub ahead of print 2016. 10.1007/s12028-016-0306-2.

16.

Kim

Song

et al.

Red blood cell distribution width is associated with poor clinical outcome in acute cerebral infarction. Thromb Haemost 2012; 108: 349–356.

17.

Tokgoz

Keskin

Kayrak

et al.

Is neutrophil/lymphocyte ratio predict to short-term mortality in acute cerebral infarct independently from infarct volume?

J Stroke Cerebrovasc Dis 2014; 23: 2163–2168.

18.

Saeed

Villarroel

Reisner

et al.

Multiparameter Intelligent Monitoring in Intensive Care II: a public-access intensive care unit database. Crit Care Med 2011; 39: 952–960.

19.

Zhang

. Accessing critical care big data: a step by step approach. J Thorac Dis 2015; 7: 238–242.

20.

Wei

Tang

et al.

Prognostic value of red blood cell distribution width in acute pancreatitis patients admitted to intensive care units: an analysis of a publicly accessible clinical database MIMIC II. Clin Chem Lab Med 2016; 54: e195–e197.

21.

Wei

Zhong

. Red blood cell distribution: an index without additional cost in estimating the prognosis of acute pancreatitis. Clin Chem Lab Med 2016; 54: e389–e390.

22.

Le Gall

Loirat

Alperovitch

et al.

A simplified acute physiology score for ICU patients. Crit Care Med 1984; 12: 975–977.

23.

Zhang

. Model building strategy for logistic regression: purposeful selection. Ann Transl Med 2016; 4: 111–111.

24.

Lippi

Plebani

. Red blood cell distribution width (RDW) and human pathology. One size fits all. Clin Chem Lab Med 2014; 52: 1247–1249.

25.

Provencio

Vora

. Subarachnoid hemorrhage and inflammation: bench to bedside and back. Semin Neurol 2005; 25: 435–444.

26.

Juvela

Kuhmonen

Siironen

. C-reactive protein as predictor for poor outcome after aneurysmal subarachnoid haemorrhage. Acta Neurochir (Wien) 2012; 154: 397–404.

27.

McMahon

Hopkins

Vail

et al.

Inflammation as a predictor for delayed cerebral ischemia after aneurysmal subarachnoid haemorrhage. J Neurointerv Surg 2013; 5: 512–517.

28.

Lattanzi

Cagnetti

Provinciali

et al.

Neutrophil-to-lymphocyte ratio predicts the outcome of acute intracerebral hemorrhage. Stroke 2016; 47: 1654–1657.

29.

Lippi

Targher

Montagnana

et al.

Relation between red blood cell distribution width and inflammatory biomarkers in a large cohort of unselected outpatients. Arch Pathol Lab Med 2009; 133: 628–632.

30.

Vaya

Sarnago

Fuster

et al.

Influence of inflammatory and lipidic parameters on red blood cell distribution width in a healthy population. Clin Hemorheol Microcirc 2015; 59: 379–385.

31.

Huang

Qin

et al.

Prognostic value of neutrophil to lymphocyte ratio for gastric cancer. Ann Transl Med 2015; 3: 50–50.

32.

Bhat

Teli

Rijal

et al.

Neutrophil to lymphocyte ratio and cardiovascular diseases: a review. Expert Rev Cardiovasc Ther 2013; 11: 55–59.

33.

Qin

et al.

Neutrophil- and platelet-to-lymphocyte ratios are correlated with disease activity in rheumatoid arthritis. Clin Lab 2015; 61: 269–273.

34.

Qin

Tang

et al.

Neutrophil to lymphocyte ratio (NLR) and platelet to lymphocyte ratio (PLR) were useful markers in assessment of inflammatory response and disease activity in SLE patients. Mod Rheumatol 2016; 26: 372–376.

35.

Gunay

Sarinc Ulasli

Akar

et al.

Neutrophil-to-lymphocyte ratio in chronic obstructive pulmonary disease: a retrospective study. Inflammation 2014; 37: 374–380.

36.

Vespa

Bleck

. Neurogenic pulmonary edema and other mechanisms of impaired oxygenation after aneurysmal subarachnoid hemorrhage. Neurocrit Care 2004; 1: 157–170.

37.

Ycas

Horrow

Horne

. Persistent increase in red cell size distribution width after acute diseases: a biomarker of hypoxemia? Clin Chim Acta 2015; 448: 107–117.

38.

Schuiling

de Weerd

Dennesen

et al.

The simplified acute physiology score to predict outcome in patients with subarachnoid hemorrhage. Neurosurgery 2005; 57: 230–26. discussion 6.

39.

Zhang

et al.

Red cell distribution width is associated with hospital mortality in unselected critically ill patients. J Thorac Dis 2013; 5: 730–736.

40.

Hunziker

Celi

Lee

et al.

Red cell distribution width improves the simplified acute physiology score for risk prediction in unselected critically ill patients. Crit Care 2012; 16: R89–R89.

41.

Fujita

Franz

Figulla

et al.

Red cell distribution width and survival in patients hospitalized on a medical ICU. Clin Biochem 2015; 48: 1048–1052.

42.

Salciccioli

Marshall

Pimentel

et al.

The association between the neutrophil-to-lymphocyte ratio and mortality in critical illness: an observational cohort study. Crit Care 2015; 19: 13–13.

43.

Akilli

Yortanli

Mutlu

et al.

Prognostic importance of neutrophil-lymphocyte ratio in critically ill patients: short- and long-term outcomes. Am J Emerg Med 2014; 32: 1476–1480.

44.

Shen

Wang

et al.

The prognostic value of plasma thrombospondin-1 concentrations after aneurysmal subarachnoid hemorrhage. Clin Chim Acta 2015; 448: 155–160.

45.

Chen

Sun

et al.

The prognostic value of plasma soluble CD40 ligand levels following aneurysmal subarachnoid hemorrhage. Thromb Res 2015; 136: 24–29.

46.

Shemin

Rittenberg

. The life span of the human red blood cell. J Biol Chem 1946; 166: 627–636.

47.

Lopponen

Qian

Tetri

et al.

Predictive value of C-reactive protein for the outcome after primary intracerebral hemorrhage. J Neurosurg 2014; 121: 1374–1379.

48.

Fang

et al.

Effects of plasma D-dimer levels on early mortality and long-term functional outcome after spontaneous intracerebral hemorrhage. J Clin Neurosci 2014; 21: 1364–1367.

49.

Buratti

Cagnetti

Balucani

et al.

Blood pressure variability and stroke outcome in patients with internal carotid artery occlusion. J Neurol Sci 2014; 339: 164–168.

50.

Lattanzi

Cagnetti

Provinciali

et al.

Blood pressure variability and clinical outcome in patients with acute intracerebral hemorrhage. J Stroke Cerebrovasc Dis 2015; 24: 1493–1499.