Accepted Manuscript Low serum TNF-related apoptosis-inducing ligand (TRAIL) levels are associated with acute ischemic stroke severity Yang Ho Kang, Min-Gyu Park, Kyung-Ha Noh, Hae Rim Park, Hye Won Lee, Seok Man Son, Kyung-Pil Park PII:
S0021-9150(15)00188-4
DOI:
10.1016/j.atherosclerosis.2015.03.028
Reference:
ATH 14004
To appear in:
Atherosclerosis
Received Date: 12 January 2015 Revised Date:
4 March 2015
Accepted Date: 18 March 2015
Please cite this article as: Kang YH, Park M-G, Noh K-H, Park HR, Lee HW, Son SM, Park K-P, Low serum TNF-related apoptosis-inducing ligand (TRAIL) levels are associated with acute ischemic stroke severity, Atherosclerosis (2015), doi: 10.1016/j.atherosclerosis.2015.03.028. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT Low serum TNF-related apoptosis-inducing ligand (TRAIL) levels are associated with acute ischemic stroke severity
RI PT
Yang Ho Kanga,c, Min-Gyu Parkb,c, Kyung-Ha Nohb,c, Hae Rim Parkb,c, Hye Won Leea,c , Seok Man Sona,c, Kyung-Pil Parkb,c,* a
Division of Endocrinology and Metabolism, Department of Internal Medicine, Pusan
SC
National University Yangsan Hospital, Pusan National University School of Medicine, Yangsan, Korea
Department of Neurology, Pusan National University Yangsan Hospital, Pusan National
M AN U
b
University School of Medicine, Yangsan, Korea c
Research Institute for Convergence of Biomedical Science and Technology, Pusan National
The number of tables: 4
EP
The number of figures: 2
TE D
University Yangsan Hospital, Yangsan, Korea
AC C
Keywords: TRAIL, acute ischemic stroke, cerebral infarction
* Corresponding author: Kyung-Pil Park, MD, PhD Department of Neurology, Pusan National University Yangsan Hospital, Pusan National University School of Medicine, 20 Geumo-ro, Mulgeum, Yangsan 626-770, Republic of Korea, tel.: +82 55 360 2451; fax: +82 55 360 2152 E-Mail address: [email protected] (K. Park)
ACCEPTED MANUSCRIPT Abstract
Background: TNF-related apoptosis-inducing ligand (TRAIL) is a member of the tumor
RI PT
necrosis factor receptor superfamily and its serum level is known to be closely associated with future cardiovascular events and prognosis of various cardiovascular diseases. We investigated whether serum TRAIL levels are associated with the severity of acute ischemic
SC
stroke and specific stroke subtype.
Methods: We used an enzyme-linked immunosorbent assay to measure the serum TRAIL
M AN U
levels of 293 patients with acute ischemic stroke within 7 days of onset. Stroke subtype was classified as large artery atherosclerosis, cardioembolism, small vessel occlusion and other determined etiology. We used National Institute of Health Stroke Scale (NIHSS) score of first hospital day and stroke volume on diffusion-weighted imaging within 7 days of stroke onset
TE D
for measuring the severity of acute ischemic stroke.
Results: The level of serum TRAIL showed significant negative correlations with NIHSS score and stroke volume. Serum TRAIL levels significantly decreased as the tertile of NIHSS
EP
score and stroke volume increased. The relative risk of patients with serum TRAIL < 64.0 pg/mL for the presence of highest tertile of NIHSS score was significantly increased
AC C
(adjusted OR, [95%CI]; 7.07 [3.64 - 13.74]). Regarding stroke volume, the relative risk of patients with serum TRAIL < 71.5 pg/mL for the presence of highest tertile of stroke volume was also significantly increased (adjusted OR, [95%CI]; 2.81 [1.61 – 4.92]). There are no significant differences of serum TRAIL level among stroke subtypes. Conclusions: Low serum TRAIL levels were significantly associated with the acute ischemic stroke severity. This finding suggests that serum TRAIL might also have a role in acute ischemic stroke as well as other cardiovascular diseases.
ACCEPTED MANUSCRIPT 1. Introduction
Tumor necrosis factor (TNF)-related apoptosis-inducing ligand (TRAIL) is a member of the
RI PT
TNF superfamily and is primarily known to involve in the process of apoptosis preferentially in cancer cells through its interaction with death receptors [1,2]. On the other hand, the possible roles of TRAIL in the cardiovascular system have not been fully elucidated. Earlier
SC
studies have shown that TRAIL has the ability to protect from apoptosis and promote proliferation of vascular endothelial cells by activating the Akt and extracellular signal-
M AN U
regulated kinase (ERK) pathways [3]. The addition of TRAIL to primary human endothelial cells increased phosphorylation of endothelial nitric oxide synthase (eNOS), NOS activity, and nitric oxide synthesis, suggesting that TRAIL might play an important role in endothelial cell function [4]. Moreover, a study has demonstrated that systemic administration of
progression [5].
TE D
recombinant TRAIL protein showed a protective role in vivo against atherosclerosis
Several clinical studies have shown that low levels of soluble TRAIL are associated with
EP
poor prognoses in patients affected by chronic heart failure [6], cardiovascular disease [7], and acute coronary syndrome (including acute myocardial infarction or unstable angina
AC C
pectoris) [8-10]. However, to the best of our knowledge, no studies have examined the association between serum TRAIL levels and acute ischemic stroke. Therefore, we investigated whether serum TRAIL levels are associated with the severity of acute ischemic stroke and specific stroke subtype.
ACCEPTED MANUSCRIPT 2. Methods
RI PT
2.1. Patients
The study participants consisted of patients with acute ischemic stroke who were admitted to the neurological department of our university hospital within 7 days after stroke onset.
SC
Acute stroke was confirmed by diffusion-weighted magnetic resonance (MR) imaging in all enrolled patients. This study was approved by the Institutional Review Board of our hospital
of Helsinki of 1975.
TE D
2.2. Biochemical measures
M AN U
and written informed content was obtained from all patients in accordance with Declaration
Blood samples were collected from all acute stroke patients within three days of stroke onset and were separated by centrifugation of the blood at 1900 g for 15 min at 4 °C and were
EP
stored at - 80 °C and thawed only once before analyses. Serum TRAIL levels were measured using a commercially available enzyme-linked
AC C
immunosorbent assay (ELISA) kit (R&D Systems, Minneapolis, MN), in accordance with the manufacturer's instructions and analyzed with an ELISA reader at 450 nm. All samples were measured in duplicate and averaged. Sensitivity of the assay was 2.86 pg/mL and the intraand inter-assay coefficients of variation (CV) were 3.9 % and 5.9 %, respectively and the upper limit of detection was 1000 pg/mL. Blood samples for biochemistry [e.g. fasting blood glucose, total cholesterol, low-density lipoprotein (LDL) cholesterol, triglyceride, highdensity lipoprotein (HDL) cholesterol, high-sensitivity C-reactive protein (hs-CRP)] and hematology (e.g. hemoglobin level, leukocyte count) were taken at admission.
ACCEPTED MANUSCRIPT 2.3. Determination of presence and severity of acute ischemic stroke
We included patients who have undergone diffusion-weighted imaging (DWI) and at least
RI PT
one angiographic study using digital subtraction angiography (DSA), MR angiography (MRA) or CT angiography (CTA) during the admission period. MRI and MRA were conducted using a 1.5 T (Avanto; Siemens, Erlangen, Germany) or 3.0 T system (Verio or Skyra; Siemens,
SC
Erlangen, Germany). We divided patients into four categories as follows: 1) large artery atherosclerosis 2) cardioembolism (CE), 3) small vessel occlusion (SVO), and 4) stroke of
M AN U
other determined etiology (OT) according to the Trial of Org 10,172 in Acute Stroke Treatment (TOAST) classification system [11]. We excluded the patients with undetermined etiology in TOAST classification.
We used the National Institute of Health Stroke Scale (NIHSS) score on admission [12] and
TE D
initial diffusion MRI lesion volume to determine the severity of acute stroke. Diffusion lesion volume was measured manually. All areas of diffusion lesion were contoured on a slice-byslice basis by manual editing. Volumes were calculated based on slice thickness and overall
EP
outlined lesion areas. All patients were grouped into tertiles according to their NIHSS scores or stroke volumes. The range of NIHSS scores for each tertile was as follows: tertile 1 (n =
AC C
135), 0 - 3 (median, 1); tertile 2 (n = 87), 4 - 7 (median, 5); and tertile 3 (n = 71), 8 - 24 (median, 14). The range of stroke volume for each tertile was as follows: tertile 1, (n = 98), 15.8 - 612.5 (median, 286.1 mm3); tertile 2, (n = 98), 614.9 – 5769.0 (median, 1512.0 mm3); and tertile 3, (n = 97), 5992.0 – 255770.0 (median, 29719.0 mm3).
2.4. Other measures
Baseline clinical data, including age, sex, and time from stroke onset to admission, were
ACCEPTED MANUSCRIPT collected for all patients. Body mass index (BMI, kg/m2) was calculated as weight in kilograms divided by height in meters squared. Data on the presence of hypertension, diabetes mellitus, hypercholesterolemia, and current smoking were also obtained.
RI PT
Hypertension was diagnosed when a patient was taking anti-hypertensive medication or had resting systolic blood pressures ≥ 140 mmHg or resting diastolic blood pressures ≥ 90 mmHg on repeated measurements [13]. Diabetes mellitus was diagnosed if the patient had fasting
SC
blood glucose ≥ 126 mg/dL or was being treated with anti-diabetic medications or insulin [14]. Hypercholesterolemia was diagnosed if the patient had LDL cholesterol ≥ 190 mg/dL or
M AN U
treated with lipid lowering agents after the diagnosis of dyslipidemia [15]. Current smoking was defined when a patient had smoked a cigarette within 1 year prior to admission.
2.5. Statistical analysis
TE D
Categorical variables were recorded as frequencies or counts (percentages). Continuous data were tested for distribution using the Kolmogorov-Smirnov test. Fisher’s exact test was used to compare categorical variables between groups. All patients were divided into tertiles based
EP
on the distribution of NIHSS and stroke volume. For continuous variables, parameters that
AC C
followed a normal distribution were analyzed with a Student’s t-test or analysis of variance (ANOVA) and described as the mean ± standard deviation (SD). Parameters that did not follow a normal distribution were analyzed with the Mann-Whitney U test or the KruskalWallis test and expressed as the median (interquartile range). Correlations between TRAIL and variables were assessed using Spearman’s correlation coefficient. Values with skewed distribution were log-transformed before statistical analysis. A receiver-operating characteristic (ROC) curve analysis was used to determine the diagnostic accuracy and optimal cut-off values of TRAIL for the severity of acute stroke
ACCEPTED MANUSCRIPT defined by NIHSS score or stroke volume. We treated the serum TRAIL level as a dichotomized variable (by the optimal cut-off) in statistical analyses. We performed multiple logistic regression analysis using the highest tertile of NIHSS score or stroke volume as
RI PT
dependent variable. We adjusted for potential confounders regarded to be clinically relevant, including the following covariates: age, sex, BMI, hypertension, diabetes mellitus, hypercholesterolemia, and current smoking. All tests were two-tailed, and a p value < 0.05
AC C
EP
TE D
M AN U
18.0 software (SPSS Inc., Chicago, Illinois)
SC
was considered statistically significant. All data analyses were performed using SPSS version
ACCEPTED MANUSCRIPT 3. Results
Two hundred and ninety three patients (men 193, women 100) were enrolled in the study.
RI PT
The mean age of the patients was 65.7 ± 11.2 years. The median serum level of TRAIL was 98.4 (70.1 - 141.0) pg/mL. The median NIHSS score on admission was 4 (2 - 7). The median stroke volume was 1506.9 mm3 (433.7 - 14108.3 mm3). There were no significant differences
SC
in median levels of serum TRAIL, NIHSS score, and stroke volume between men and women (Table 1). There were no significant differences in serum TRAIL levels between patients with
M AN U
and without hypertension, diabetes mellitus, dyslipidemia, and current smoking. Also there were no significant differences in serum TRAIL levels among stroke subtype by TOAST classification (Table 2).
In age-adjusted Spearman's correlation coefficients, log (TRAIL) was negatively correlated
TE D
with fasting blood glucose (FBG) (r = -0.132, P = 0.035), log (NIHSS score) (r = -0.249, P < 0.001), and log (stroke volume) (r = -0.252, P < 0.001). Age also showed a negative correlation with log (TRAIL), but it did not reach statistical significance (r = -0.115, P =
EP
0.065). On the other hand, log (TRAIL) showed weak positive correlations with BMI (r = 0.178, P = 0.004), total cholesterol (r = 0.152, P = 0.015), and LDL cholesterol (r = 0.145, P
AC C
= 0.020) (Table 3).
The serum TRAIL level significantly decreased as the tertile of NIHSS score (tertile 1; 107.7 [36.8 - 316.8] pg/mL, tertile 2; 103.6 [29.4 - 438.2] pg/mL, and tertile 3; 66.9 [19.9 346.3] pg/mL, P < 0.05, respectively) and stroke volume (tertile 1; 113.0 [25.0 - 415.0] pg/mL, tertile 2; 97.5 [33.0 - 438.2] pg/mL, and tertile 3; 83.9 [19.9 - 346.3] pg/mL, P < 0.05, respectively) increased (Fig. 1). The receiver-operating characteristic (ROC) curve analysis showed that the area under curve
ACCEPTED MANUSCRIPT (AUC) value for TRAIL that discriminate the presence of the third tertile of NIHSS score was 0.695 [95% CI, 0.617 - 0.773]. The optimal cut-off value of TRAIL for the third tertile of NIHSS score was 64.0 pg/mL. Sensitivity and specificity at the cut-off value < 64.0 pg/mL
RI PT
were 0.451 and 0.892, respectively (Fig. 2A). On the other hand, the AUC value that discriminate the presence of the third tertile of stroke volume was 0.634 [95% CI, 0.565 0.704] and the optimal cut-off value of TRAIL for the third tertile of stroke volume was 71.5
SC
pg/mL. Sensitivity and specificity at the cut-off value < 71.5 pg/mL were 0.402 and 0.816, respectively (Fig. 2B).
M AN U
In multiple logistic regression analyses, the relative risk of patients with serum TRAIL < 64.0 pg/mL (n = 71) for the presence of the third tertile of NIHSS score was significantly increased compared to those with serum TRAIL ≥ 64.0 pg/mL (adjusted OR [95%CI], 7.07 [3.64 - 13.74], p < 0.001). Regarding stroke volume, the relative risk of patients with serum
TE D
TRAIL < 71.5 pg/mL (n = 97) for the presence of the third tertile of stroke volume was also significantly increased compared to those with serum TRAIL ≥ 71.5 pg/mL (adjusted OR
AC C
EP
[95%CI] was 2.81 [1.61 – 4.92], p < 0.001, Table 4).
ACCEPTED MANUSCRIPT 4. Discussion
In this study, we demonstrated that decreased serum TRAIL levels were associated with the
RI PT
severity of acute ischemic stroke reflected by the NIHSS score and stroke lesion volume on admission. To our knowledge, this is the first report to evaluate the association of serum TRAIL levels with acute ischemic stroke. Several cross sectional and prospective clinical
SC
studies have shown that low serum TRAIL levels are associated with a higher risk for death and worse clinical outcome in patients with various cardiovascular diseases such as chronic
M AN U
heart failure and acute coronary syndrome [6-10]. Our findings suggest that serum TRAIL has important role of apoptosis related brain cell death in acute ischemic stroke. The possible function and role of TRAIL in the cardiovascular system is not completely elucidated. TRAIL leads to apoptotic cell death in tumor cells according to the in vitro studies
TE D
[16,17] and one study reported that soluble TRAIL can kill the tumor cells selectively without apparent toxicity to normal cells in mice [18]. However, there have been conflicting reports for the role of TRAIL in vasculature. Li et al. have shown that soluble TRAIL also induces
EP
apoptosis and pro-inflammatory activity in normal human umbilical vein endothelial cells [19]. On the other hand, several studies have demonstrated possible anti-inflammatory and
AC C
anti-atherogenic activity in vitro and in vivo [4,5,20]. The results of previous clinical studies addressing the relationship between low serum TRAIL and cardiovascular diseases can be explained by the anti-inflammatory property of TRAIL [6-10]. Furthermore, our findings of low serum TRAIL level with the increase of acute ischemic stroke severity also could be additional data, which support the anti-inflammatory role of TRAIL. Osmancik et al. reported that low concentrations of soluble TRAIL represent a strong predictor of a poor prognosis in patients with acute coronary syndrome with 100% ischemic etiology and suggested that TRAIL might be more specific for patients with ischemic
ACCEPTED MANUSCRIPT etiology of left ventricular dysfunction relative to other etiologies [10]. Consistent with a study of Osmancik et al., all patients enrolled our study were also with 100% ischemic etiology of acute stroke regardless of stroke subtype and in agreement with the possible
differences of serum TRAIL levels among stroke subtypes.
RI PT
specificity of TRAIL with ischemic etiology. In addition, there were no significant
The NIHSS is known as a valid, reliable, and reproducible neurologic severity scale [12]
SC
and the NIHSS scores on admission have an association with chronic functional outcome [21] and hospital disposition after stroke [22]. In this study, patients were divided into tertiles
M AN U
according to the NIHSS scores on admission and AUC value for TRAIL that discriminate the presence of highest tertile of NIHSS score was 0.695 [95% CI, 0.617 - 0.773]. We evaluated stroke volume as another parameter for reflecting the severity of acute ischemic stroke. When patients were grouped into the tertiles of stroke volume, AUC value for TRAIL that
TE D
discriminate the presence of highest tertile of stroke volume was somewhat lower than that of NIHSS score (95% CI, 0.634 [0.565 - 0.704]). Volpato et al. have reported that the risk of all cause and cardiovascular disease mortality linearly increases as TRAIL levels decrease near
EP
the value of 70 pg/mL, and after this level there is a curvilinear relationship between TRAIL and the risk of death, with very high mortality risk associated with lowest TRAIL levels in
AC C
1,282 adults aged 65 years or older with pre-existing cardiovascular disease [7]. In our study, the optimal cut-off value of TRAIL for highest tertile of NIHSS score used as a representative of acute ischemic stroke severity was 64.0 pg/mL and 71.5 pg/mL regarding stroke volume, which was quite close to the value of 70 pg/mL in the previous study [7]. Furthermore, Volpato et al. have demonstrated that patients with lowest quartile TRAIL concentrations have a three-fold increased risk of death, compared to those with highest TRAIL levels (Hazard ratios, 3.0; 95% CI, 1.4 - 6.5) among those with prevalent cardiovascular disease [7]. The relative risk of patients with low serum TRAIL for an increase
ACCEPTED MANUSCRIPT of the acute ischemic stroke severity was also significantly higher in our study. This study had some limitations. First, the determination of serum TRAIL levels was based on the measurement of a single sample obtained upon admission and, therefore, we could not
RI PT
evaluate possible changes of serum TRAIL according to the stages of acute ischemic stroke during treatment. Secchiero et al. have shown that serum TRAIL levels were significantly decreased in the early time points after acute myocardial infarction and progressively
SC
increased in the follow-up [23]. With regards to the acute ischemic stroke, reduced serum TRAIL levels at acute stage may be elevated as the recovery of stroke in the follow-up period,
M AN U
which need a further clinical study. Second, the correlation of TRAIL with the acute ischemic stroke severity was weak (r = 0.249 with NIHSS score and r = 0.252 with stroke volume) and the discrimination ability of TRAIL for the ischemic stroke severity was also moderate (AUC, 0.695 using NIHSS score and 0.634 using stroke volume). These results may be related with
TE D
small number of enrolled patients in our study, which limited the statistical power to evaluate the diagnostic value of serum TRAIL for the acute ischemic stroke severity. Finally, the study design was cross-sectional and, as such, causal and longitudinal relationships were not
EP
addressed.
In conclusion, our study has demonstrated that low serum TRAIL levels were associated
AC C
with the severity of acute ischemic stroke for the first time and it does provide additional information regarding relationship between low serum TRAIL level and poor prognosis or outcome of cardiovascular disease. Our findings suggest that serum TRAIL also could be a potential marker for the acute ischemic stroke as well as other cardiovascular diseases.
ACCEPTED MANUSCRIPT Sources of funding This work was supported by Pusan National University Research Grant, 2013.
AC C
EP
TE D
M AN U
SC
Authors had no financial conflicts with commercial entities.
RI PT
Disclosures
ACCEPTED MANUSCRIPT Table 1. Clinical characteristics of patients. Total
Men
Women
Number
293
193
100
Age, years
65.7 ± 11.2
63.8 ± 10.7
69.2 ± 11.3
BMI, kg/m2
23.4 ± 3.0
23.6 ± 2.9
FBG, mg/dL
119.3 ± 28.0
117.8 ± 28.3
Cholesterol, mg/dL
187.2 ± 40.7
187.3 ± 40.5
LDL-C, mg/dL
115.9 ± 36.1
116.2 ± 34.7
115.3 ± 38.6
0.839a
TG, mg/dL
99.0 (73.0 - 142.5)
104.0 (79.0 - 149.0)
85.0 (62.0 - 122.7)
< 0.001b
HDL-C, mg/dL
45.7 ± 13.3
45.0 ± 13.6
45.7 ± 12.7
0.626a
hs-CRP, mg/L
0.15 (0.06 - 0.36)
0.15 (0.07 - 0.39)
0.12 (0.05 - 0.32)
0.178b
Hemoglobin, g/dL
14.4 ± 7.6
15.3 ± 9.2
12.7 ± 1.7
< 0.001a
WBC, ×109/L
7.8 ± 2.9
8.1 ± 3.1
7.5 ± 2.4
0.070a
TRAIL, pg/mL
98.4 (70.1 - 141.0)
100.1 (73.9 - 150.0)
91.5 (64.2 - 129.7)
0.120b
NIHSS score
4 (2 - 7)
4 (2 - 7)
4 (2 - 9)
0.174b
Stroke Volume,
1506.9
1517.0
1477.0
0.845b
(433.7 - 14108.3)
(433.1 - 16144.3)
(433.9 - 12485.1)
191 (65.2%)
120 (62.2%)
71 (71.0%)
0.155c
AC C
Hypertension
P value*
< 0.001a
23.0 ± 3.3
0.157a
121.9 ± 27.3
0.230a
186.9 ± 41.2
0.943a
SC
M AN U
TE D
EP
mm3/L
RI PT
Characteristic
Diabetes mellitus
94 (32.1%)
61 (31.6%)
32 (32.0%)
1.000c
Dyslipidemia
48 (16.4%)
28 (14.5%)
20 (20.0%)
0.246c
Current smoking
140 (47.8%)
132 (68.4%)
8 (0.8%)
< 0.001c
CE 65 (22.2%)
37 (19.2%)
28 (28.0%)
0.103 c
LAA 83 (28.3%)
66 (34.2%)
17 (17.0%)
0.002 c
Stroke subtype
ACCEPTED MANUSCRIPT SVO 95 (32.4%)
60 (31.1%)
35 (35.0%)
0.513 c
OT 50 (17.1%)
30 (15.5%)
20 (20.0%)
0.332 c
Continuous variables are expressed as mean ± standard deviation (SD) or median
RI PT
(interquartile range). The categorical values are given as frequencies and respective percentages.
Abbreviations: BMI, body mass index; FBG, fasting blood glucose; LDL-C, low-density
SC
lipoprotein cholesterol; TG, triglyceride; HDL-C, high-density lipoprotein cholesterol; hsCRP, high-sensitivity C-reactive protein; WBC, white blood cell counts; TRAIL, tumor
M AN U
necrosis factor-related apoptosis inducing ligand; NIHSS, National Institute of Health Stroke Scale; CE, cardioembolism; LAA, large artery atherosclerosis; SVO, small vessel occlusion, OT, stroke of other determined etiology.
Student’s t-test.
b
Mann-Whitney U test.
c
Fisher’s exact test.
AC C
EP
a
TE D
*The P value was calculated by a comparison between men and women.
ACCEPTED MANUSCRIPT Table 2. Relationship between serum TRAIL and risk factors. Parameter
TRAIL, pg/mL
Yes (191)
100.1 (73.1- 438.2)
No (102)
87.8 (19.9 - 346.3)
Yes (94)
99.3 (25.0 - 438.2)
No (199)
96.4 (19.9 - 415.0)
0.281a
M AN U
Dyslipidemia (n) Yes (48)
109.6 (29.4 - 340.0)
No (245)
96.2 (19.9 - 438.2)
0.671a
Yes (140)
98.5 (19.9 - 346.3)
No (153)
TE D
Current smoking (n)
96.5 (25.0 - 438.2)
Stroke subtype (n)
0.045b
85.5 (27.4 - 316.0)
AC C
EP
CE (65)
SVO (95)
SC
0.478a
Diabetes Mellitus (n)
RI PT
0.130a
Hypertension (n)
LAA (83)
P value
OT (50)
102.6 (34.0 - 346.3) 107.7 (34.7 - 438.2) 92.3 (19.9 - 305.0)
Values are expressed as median (interquartile range). Abbreviations: CE, cardioembolism; LAA, large artery atherosclerosis; SVO, small vessel occlusion, OT, stroke of other determined etiology. a
Mann-Whitney U test.
b
Kruskal-Wallis test.
ACCEPTED MANUSCRIPT Table 3. Correlations between log (TRAIL) concentration and various parameters. P value
-0.115
0.065
BMI
0.178
0.004
FBG
-0.132
0.035
Cholesterol
0.152
0.015
LDL-C
0.145
0.020
Log (TG)
0.120
0.054
HDL-C
0.099
0.112
Log (hs-CRP)
-0.102
0.103
Log (NIHSS score)
-0.249
< 0.001
Log (Stroke Volume)
-0.252
< 0.001
M AN U
Age
RI PT
r
SC
Parameters
TE D
Abbreviations: BMI, body mass index; FBG, fasting blood glucose; LDL-C, low-density lipoprotein cholesterol; TG, triglyceride; HDL-C, high-density lipoprotein cholesterol; hs-
AC C
EP
CRP, high-sensitivity C-reactive protein; NIHSS, National Institute of Health Stroke Scale.
ACCEPTED MANUSCRIPT Table 4. Relationships between serum TRAIL and stroke severity. For 3T of NIHSS score
For 3T of stroke volume
TRAIL < 64.0 pg/mL
TRAIL < 71.5 pg/mL
95% CI
P value
ORb
95% CI
P value
7.07
[3.64-13.74] < 0.001
2.81
[1.61-4.92]
< 0.001
RI PT
ORa
Dependent is the presence of the third tertile of NIHSS score or stroke volume.
Odds ratio for patients with TRAIL < 64.0 pg/mL compared to those with TRAIL ≥ 64.0
SC
a
pg/mL.
Odds ratio for patients with TRAIL < 71.5 pg/mL compared to those with TRAIL ≥ 71.5
pg/mL. Adjusted
for
age,
sex,
body
hypercholesterolemia, current smoking.
M AN U
b
mass
index,
hypertension,
diabetic
mellitus,
TE D
Abbreviations: 3T, tertile 3; NIHSS, National Institute of Health Stroke Scale; TRAIL, tumor
AC C
EP
necrosis factor-related apoptosis inducing ligand; OR, odds ratio; CI, confidence interval.
RI PT
ACCEPTED MANUSCRIPT
SC
Fig. 1. Serum TRAIL levels according to the tertiles of NIHSS score (A) and stroke volume (B). Serum TRAIL levels were negatively associated with the severity of acute ischemic
M AN U
stroke.
In each box plot, the lower and upper ends of the box represent the 25th and 75th percentiles (interquartile range, IQR) and the line inside the box represents the median of serum TRAIL (pg/mL). The peripheral lines extend to the outer fences, which are 1.5 times the interquartile
TE D
range below and above the 25th and 75th percentile, respectively.
Abbreviations: NIHSS, National Institute of Health Stroke Scale Score; 1T, tertile 1; 2T, tertile 2; 3T, tertile 3.
EP
Kruskal-Wallis test with Post hoc (Dunn's Multiple Comparison Test) test.
AC C
*, P < 0.05 vs. 1T or 2T.
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
Fig. 2. Receiver-operating characteristic curve for TRAIL to discriminate the severity of acute ischemic stroke. The area under curve (AUC) for the presence of the highest tertile of NIHSS score (A) and the highest tertile of stroke volume (B).
Abbreviations: TRAIL, tumor necrosis factor-related apoptosis inducing ligand; NIHSS,
TE D
National Institute of Health Stroke Scale; AUC, area under the receiver-operating
AC C
EP
characteristic curve; CI, confidence interval.
ACCEPTED MANUSCRIPT References [1]
Manzo F, Nebbioso A, Miceli M, Conte M, De Bellis F, Carafa V, et al. TNF-related apoptosis-inducing ligand: Signalling of a “smart” molecule. The International
[2]
RI PT
Journal of Biochemistry & Cell Biology 41 (2009) 460–466. Wiley SR, Schooley K, Smolak PJ, Din WS, Huang CP, Nicholl JK, et al. Identification and characterization of a new member of the TNF family that induces
[3]
SC
apoptosis. Immunity 3 (1995) 673–682.
Secchiero P, Gonelli A, Carnevale E, Milani D, Pandolfi A, Zella D, et al. TRAIL
M AN U
promotes the survival and proliferation of primary human vascular endothelial cells by activating the Akt and ERK pathways. Circulation 107 (2003) 2250–2256. [4]
Zauli G, Pandolfi A, Gonelli A, Di Pietro R, Guarnieri S, Ciabattoni G, et al. Tumor necrosis factor-related apoptosis-inducing ligand (TRAIL) sequentially upregulates
TE D
nitric oxide and prostanoid production in primary human endothelial cells. Circulation Research 92 (2003) 732–740. [5]
Secchiero P, Candido R, Corallini F, Zacchigna S, Toffoli B, Rimondi E, et al.
EP
Systemic tumor necrosis factor-related apoptosis-inducing ligand delivery shows antiatherosclerotic activity in apolipoprotein E-null diabetic mice. Circulation 114
[6]
AC C
(2006) 1522–1530.
Niessner A, Hohensinner PJ, Rychli K, Neuhold S, Zorn G, Richter B, et al.
Prognostic value of apoptosis markers in advanced heart failure patients. European Heart Journal 30 (2008) 789–796.
[7]
Volpato S, Ferrucci L, Secchiero P, Corallini F, Zuliani G, Fellin R, et al. Association of tumor necrosis factor-related apoptosis-inducing ligand with total and cardiovascular mortality in older adults. Atherosclerosis 215 (2011) 452–458.
[8]
Schoppet M, Sattler AM, Schaefer JR, Hofbauer LC. Osteoprotegerin (OPG) and
ACCEPTED MANUSCRIPT tumor necrosis factor-related apoptosis-inducing ligand (TRAIL) levels in atherosclerosis. Atherosclerosis 184 (2006) 446–447. [9]
Michowitz Y, Goldstein E, Roth A, Afek A, Abashidze A, Ben Gal Y, et al. The
RI PT
involvement of tumor necrosis factor-related apoptosis-inducing ligand (TRAIL) in atherosclerosis. Journal of the American College of Cardiology 45 (2005) 1018–1024. [10]
Osmancik P, Teringova E, Tousek P, Paulu P, Widimsky P. Prognostic Value of
Patients. PLoS ONE 8 (2013) e53860.
Adams HP, Bendixen BH, Kappelle LJ, Biller J, Love BB, Gordon DL, et al.
M AN U
[11]
SC
TNF-Related Apoptosis Inducing Ligand (TRAIL) in Acute Coronary Syndrome
Classification of subtype of acute ischemic stroke. Definitions for use in a multicenter clinical trial. TOAST. Trial of Org 10172 in Acute Stroke Treatment. Stroke 24 (1993) 35–41.
Kasner SE. Clinical interpretation and use of stroke scales. Lancet Neurol 5 (2006)
TE D
[12]
603–612. [13]
Mancia G, De Backer G, Dominiczak A, Cifkova R, Fagard R, Germano G, et al.
EP
2007 Guidelines for the Management of Arterial Hypertension: The Task Force for the Management of Arterial Hypertension of the European Society of Hypertension
AC C
(ESH) and of the European Society of Cardiology (ESC). J Hypertens 25 (2007) 1105–1187.
[14]
American Diabetes Association. Standards of medical care in diabetes--2014.
Diabetes Care 37 (2014) Suppl 1:S14–80.
[15]
Stone NJ, Robinson JG, Lichtenstein AH, Bairey Merz CN, Blum CB, Eckel RH, et al. 2013 ACC/AHA guideline on the treatment of blood cholesterol to reduce atherosclerotic cardiovascular risk in adults: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines. Journal
ACCEPTED MANUSCRIPT of the American College of Cardiology 63 (2014) 2889–2934. [16]
Kischkel FC, Lawrence DA, Chuntharapai A, Schow P, Kim KJ, Ashkenazi A. Apo2L/TRAIL-dependent recruitment of endogenous FADD and caspase-8 to death
[17]
RI PT
receptors 4 and 5. Immunity 12 (2000) 611–620. Sprick MR, Weigand MA, Rieser E, Rauch CT, Juo P, Blenis J, et al. FADD/MORT1 and caspase-8 are recruited to TRAIL receptors 1 and 2 and are
[18]
SC
essential for apoptosis mediated by TRAIL receptor 2. Immunity 12 (2000) 599–609. Walczak H, Miller RE, Ariail K, Gliniak B, Griffith TS, Kubin M, et al. Tumoricidal
5 (1999) 157–163. [19]
M AN U
activity of tumor necrosis factor-related apoptosis-inducing ligand in vivo. Nat Med
Li JH, Kirkiles-Smith NC, McNiff JM, Pober JS. TRAIL induces apoptosis and inflammatory gene expression in human endothelial cells. The Journal of
[20]
TE D
Immunology 171 (2003) 1526–1533.
Secchiero P. TRAIL counteracts the proadhesive activity of inflammatory cytokines in endothelial cells by down-modulating CCL8 and CXCL10 chemokine expression
[21]
EP
and release. Blood 105 (2005) 3413–3419. Frankel MR, Morgenstern LB, Kwiatkowski T, Lu M, Tilley BC, Broderick JP, et al.
AC C
Predicting prognosis after stroke: a placebo group analysis from the National Institute of Neurological Disorders and Stroke rt-PA Stroke Trial. Neurology 55 (2000) 952–959.
[22]
Schlegel DJ, Tanne D, Demchuk AM, Levine SR, Kasner SE, Multicenter rt-PA
Stroke Survey Group. Prediction of hospital disposition after thrombolysis for acute ischemic stroke using the National Institutes of Health Stroke Scale. Arch Neurol 61 (2004) 1061–1064. [23]
Secchiero P, Corallini F, Ceconi C, Parrinello G, Volpato S, Ferrari R, et al. Potential
ACCEPTED MANUSCRIPT Prognostic Significance of Decreased Serum Levels of TRAIL after Acute
AC C
EP
TE D
M AN U
SC
RI PT
Myocardial Infarction. PLoS ONE 4 (2009) e4442.
S0021-9150(15)00188-4
DOI:
10.1016/j.atherosclerosis.2015.03.028
Reference:
ATH 14004
To appear in:
Atherosclerosis
Received Date: 12 January 2015 Revised Date:
4 March 2015
Accepted Date: 18 March 2015
Please cite this article as: Kang YH, Park M-G, Noh K-H, Park HR, Lee HW, Son SM, Park K-P, Low serum TNF-related apoptosis-inducing ligand (TRAIL) levels are associated with acute ischemic stroke severity, Atherosclerosis (2015), doi: 10.1016/j.atherosclerosis.2015.03.028. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT Low serum TNF-related apoptosis-inducing ligand (TRAIL) levels are associated with acute ischemic stroke severity
RI PT
Yang Ho Kanga,c, Min-Gyu Parkb,c, Kyung-Ha Nohb,c, Hae Rim Parkb,c, Hye Won Leea,c , Seok Man Sona,c, Kyung-Pil Parkb,c,* a
Division of Endocrinology and Metabolism, Department of Internal Medicine, Pusan
SC
National University Yangsan Hospital, Pusan National University School of Medicine, Yangsan, Korea
Department of Neurology, Pusan National University Yangsan Hospital, Pusan National
M AN U
b
University School of Medicine, Yangsan, Korea c
Research Institute for Convergence of Biomedical Science and Technology, Pusan National
The number of tables: 4
EP
The number of figures: 2
TE D
University Yangsan Hospital, Yangsan, Korea
AC C
Keywords: TRAIL, acute ischemic stroke, cerebral infarction
* Corresponding author: Kyung-Pil Park, MD, PhD Department of Neurology, Pusan National University Yangsan Hospital, Pusan National University School of Medicine, 20 Geumo-ro, Mulgeum, Yangsan 626-770, Republic of Korea, tel.: +82 55 360 2451; fax: +82 55 360 2152 E-Mail address: [email protected] (K. Park)
ACCEPTED MANUSCRIPT Abstract
Background: TNF-related apoptosis-inducing ligand (TRAIL) is a member of the tumor
RI PT
necrosis factor receptor superfamily and its serum level is known to be closely associated with future cardiovascular events and prognosis of various cardiovascular diseases. We investigated whether serum TRAIL levels are associated with the severity of acute ischemic
SC
stroke and specific stroke subtype.
Methods: We used an enzyme-linked immunosorbent assay to measure the serum TRAIL
M AN U
levels of 293 patients with acute ischemic stroke within 7 days of onset. Stroke subtype was classified as large artery atherosclerosis, cardioembolism, small vessel occlusion and other determined etiology. We used National Institute of Health Stroke Scale (NIHSS) score of first hospital day and stroke volume on diffusion-weighted imaging within 7 days of stroke onset
TE D
for measuring the severity of acute ischemic stroke.
Results: The level of serum TRAIL showed significant negative correlations with NIHSS score and stroke volume. Serum TRAIL levels significantly decreased as the tertile of NIHSS
EP
score and stroke volume increased. The relative risk of patients with serum TRAIL < 64.0 pg/mL for the presence of highest tertile of NIHSS score was significantly increased
AC C
(adjusted OR, [95%CI]; 7.07 [3.64 - 13.74]). Regarding stroke volume, the relative risk of patients with serum TRAIL < 71.5 pg/mL for the presence of highest tertile of stroke volume was also significantly increased (adjusted OR, [95%CI]; 2.81 [1.61 – 4.92]). There are no significant differences of serum TRAIL level among stroke subtypes. Conclusions: Low serum TRAIL levels were significantly associated with the acute ischemic stroke severity. This finding suggests that serum TRAIL might also have a role in acute ischemic stroke as well as other cardiovascular diseases.
ACCEPTED MANUSCRIPT 1. Introduction
Tumor necrosis factor (TNF)-related apoptosis-inducing ligand (TRAIL) is a member of the
RI PT
TNF superfamily and is primarily known to involve in the process of apoptosis preferentially in cancer cells through its interaction with death receptors [1,2]. On the other hand, the possible roles of TRAIL in the cardiovascular system have not been fully elucidated. Earlier
SC
studies have shown that TRAIL has the ability to protect from apoptosis and promote proliferation of vascular endothelial cells by activating the Akt and extracellular signal-
M AN U
regulated kinase (ERK) pathways [3]. The addition of TRAIL to primary human endothelial cells increased phosphorylation of endothelial nitric oxide synthase (eNOS), NOS activity, and nitric oxide synthesis, suggesting that TRAIL might play an important role in endothelial cell function [4]. Moreover, a study has demonstrated that systemic administration of
progression [5].
TE D
recombinant TRAIL protein showed a protective role in vivo against atherosclerosis
Several clinical studies have shown that low levels of soluble TRAIL are associated with
EP
poor prognoses in patients affected by chronic heart failure [6], cardiovascular disease [7], and acute coronary syndrome (including acute myocardial infarction or unstable angina
AC C
pectoris) [8-10]. However, to the best of our knowledge, no studies have examined the association between serum TRAIL levels and acute ischemic stroke. Therefore, we investigated whether serum TRAIL levels are associated with the severity of acute ischemic stroke and specific stroke subtype.
ACCEPTED MANUSCRIPT 2. Methods
RI PT
2.1. Patients
The study participants consisted of patients with acute ischemic stroke who were admitted to the neurological department of our university hospital within 7 days after stroke onset.
SC
Acute stroke was confirmed by diffusion-weighted magnetic resonance (MR) imaging in all enrolled patients. This study was approved by the Institutional Review Board of our hospital
of Helsinki of 1975.
TE D
2.2. Biochemical measures
M AN U
and written informed content was obtained from all patients in accordance with Declaration
Blood samples were collected from all acute stroke patients within three days of stroke onset and were separated by centrifugation of the blood at 1900 g for 15 min at 4 °C and were
EP
stored at - 80 °C and thawed only once before analyses. Serum TRAIL levels were measured using a commercially available enzyme-linked
AC C
immunosorbent assay (ELISA) kit (R&D Systems, Minneapolis, MN), in accordance with the manufacturer's instructions and analyzed with an ELISA reader at 450 nm. All samples were measured in duplicate and averaged. Sensitivity of the assay was 2.86 pg/mL and the intraand inter-assay coefficients of variation (CV) were 3.9 % and 5.9 %, respectively and the upper limit of detection was 1000 pg/mL. Blood samples for biochemistry [e.g. fasting blood glucose, total cholesterol, low-density lipoprotein (LDL) cholesterol, triglyceride, highdensity lipoprotein (HDL) cholesterol, high-sensitivity C-reactive protein (hs-CRP)] and hematology (e.g. hemoglobin level, leukocyte count) were taken at admission.
ACCEPTED MANUSCRIPT 2.3. Determination of presence and severity of acute ischemic stroke
We included patients who have undergone diffusion-weighted imaging (DWI) and at least
RI PT
one angiographic study using digital subtraction angiography (DSA), MR angiography (MRA) or CT angiography (CTA) during the admission period. MRI and MRA were conducted using a 1.5 T (Avanto; Siemens, Erlangen, Germany) or 3.0 T system (Verio or Skyra; Siemens,
SC
Erlangen, Germany). We divided patients into four categories as follows: 1) large artery atherosclerosis 2) cardioembolism (CE), 3) small vessel occlusion (SVO), and 4) stroke of
M AN U
other determined etiology (OT) according to the Trial of Org 10,172 in Acute Stroke Treatment (TOAST) classification system [11]. We excluded the patients with undetermined etiology in TOAST classification.
We used the National Institute of Health Stroke Scale (NIHSS) score on admission [12] and
TE D
initial diffusion MRI lesion volume to determine the severity of acute stroke. Diffusion lesion volume was measured manually. All areas of diffusion lesion were contoured on a slice-byslice basis by manual editing. Volumes were calculated based on slice thickness and overall
EP
outlined lesion areas. All patients were grouped into tertiles according to their NIHSS scores or stroke volumes. The range of NIHSS scores for each tertile was as follows: tertile 1 (n =
AC C
135), 0 - 3 (median, 1); tertile 2 (n = 87), 4 - 7 (median, 5); and tertile 3 (n = 71), 8 - 24 (median, 14). The range of stroke volume for each tertile was as follows: tertile 1, (n = 98), 15.8 - 612.5 (median, 286.1 mm3); tertile 2, (n = 98), 614.9 – 5769.0 (median, 1512.0 mm3); and tertile 3, (n = 97), 5992.0 – 255770.0 (median, 29719.0 mm3).
2.4. Other measures
Baseline clinical data, including age, sex, and time from stroke onset to admission, were
ACCEPTED MANUSCRIPT collected for all patients. Body mass index (BMI, kg/m2) was calculated as weight in kilograms divided by height in meters squared. Data on the presence of hypertension, diabetes mellitus, hypercholesterolemia, and current smoking were also obtained.
RI PT
Hypertension was diagnosed when a patient was taking anti-hypertensive medication or had resting systolic blood pressures ≥ 140 mmHg or resting diastolic blood pressures ≥ 90 mmHg on repeated measurements [13]. Diabetes mellitus was diagnosed if the patient had fasting
SC
blood glucose ≥ 126 mg/dL or was being treated with anti-diabetic medications or insulin [14]. Hypercholesterolemia was diagnosed if the patient had LDL cholesterol ≥ 190 mg/dL or
M AN U
treated with lipid lowering agents after the diagnosis of dyslipidemia [15]. Current smoking was defined when a patient had smoked a cigarette within 1 year prior to admission.
2.5. Statistical analysis
TE D
Categorical variables were recorded as frequencies or counts (percentages). Continuous data were tested for distribution using the Kolmogorov-Smirnov test. Fisher’s exact test was used to compare categorical variables between groups. All patients were divided into tertiles based
EP
on the distribution of NIHSS and stroke volume. For continuous variables, parameters that
AC C
followed a normal distribution were analyzed with a Student’s t-test or analysis of variance (ANOVA) and described as the mean ± standard deviation (SD). Parameters that did not follow a normal distribution were analyzed with the Mann-Whitney U test or the KruskalWallis test and expressed as the median (interquartile range). Correlations between TRAIL and variables were assessed using Spearman’s correlation coefficient. Values with skewed distribution were log-transformed before statistical analysis. A receiver-operating characteristic (ROC) curve analysis was used to determine the diagnostic accuracy and optimal cut-off values of TRAIL for the severity of acute stroke
ACCEPTED MANUSCRIPT defined by NIHSS score or stroke volume. We treated the serum TRAIL level as a dichotomized variable (by the optimal cut-off) in statistical analyses. We performed multiple logistic regression analysis using the highest tertile of NIHSS score or stroke volume as
RI PT
dependent variable. We adjusted for potential confounders regarded to be clinically relevant, including the following covariates: age, sex, BMI, hypertension, diabetes mellitus, hypercholesterolemia, and current smoking. All tests were two-tailed, and a p value < 0.05
AC C
EP
TE D
M AN U
18.0 software (SPSS Inc., Chicago, Illinois)
SC
was considered statistically significant. All data analyses were performed using SPSS version
ACCEPTED MANUSCRIPT 3. Results
Two hundred and ninety three patients (men 193, women 100) were enrolled in the study.
RI PT
The mean age of the patients was 65.7 ± 11.2 years. The median serum level of TRAIL was 98.4 (70.1 - 141.0) pg/mL. The median NIHSS score on admission was 4 (2 - 7). The median stroke volume was 1506.9 mm3 (433.7 - 14108.3 mm3). There were no significant differences
SC
in median levels of serum TRAIL, NIHSS score, and stroke volume between men and women (Table 1). There were no significant differences in serum TRAIL levels between patients with
M AN U
and without hypertension, diabetes mellitus, dyslipidemia, and current smoking. Also there were no significant differences in serum TRAIL levels among stroke subtype by TOAST classification (Table 2).
In age-adjusted Spearman's correlation coefficients, log (TRAIL) was negatively correlated
TE D
with fasting blood glucose (FBG) (r = -0.132, P = 0.035), log (NIHSS score) (r = -0.249, P < 0.001), and log (stroke volume) (r = -0.252, P < 0.001). Age also showed a negative correlation with log (TRAIL), but it did not reach statistical significance (r = -0.115, P =
EP
0.065). On the other hand, log (TRAIL) showed weak positive correlations with BMI (r = 0.178, P = 0.004), total cholesterol (r = 0.152, P = 0.015), and LDL cholesterol (r = 0.145, P
AC C
= 0.020) (Table 3).
The serum TRAIL level significantly decreased as the tertile of NIHSS score (tertile 1; 107.7 [36.8 - 316.8] pg/mL, tertile 2; 103.6 [29.4 - 438.2] pg/mL, and tertile 3; 66.9 [19.9 346.3] pg/mL, P < 0.05, respectively) and stroke volume (tertile 1; 113.0 [25.0 - 415.0] pg/mL, tertile 2; 97.5 [33.0 - 438.2] pg/mL, and tertile 3; 83.9 [19.9 - 346.3] pg/mL, P < 0.05, respectively) increased (Fig. 1). The receiver-operating characteristic (ROC) curve analysis showed that the area under curve
ACCEPTED MANUSCRIPT (AUC) value for TRAIL that discriminate the presence of the third tertile of NIHSS score was 0.695 [95% CI, 0.617 - 0.773]. The optimal cut-off value of TRAIL for the third tertile of NIHSS score was 64.0 pg/mL. Sensitivity and specificity at the cut-off value < 64.0 pg/mL
RI PT
were 0.451 and 0.892, respectively (Fig. 2A). On the other hand, the AUC value that discriminate the presence of the third tertile of stroke volume was 0.634 [95% CI, 0.565 0.704] and the optimal cut-off value of TRAIL for the third tertile of stroke volume was 71.5
SC
pg/mL. Sensitivity and specificity at the cut-off value < 71.5 pg/mL were 0.402 and 0.816, respectively (Fig. 2B).
M AN U
In multiple logistic regression analyses, the relative risk of patients with serum TRAIL < 64.0 pg/mL (n = 71) for the presence of the third tertile of NIHSS score was significantly increased compared to those with serum TRAIL ≥ 64.0 pg/mL (adjusted OR [95%CI], 7.07 [3.64 - 13.74], p < 0.001). Regarding stroke volume, the relative risk of patients with serum
TE D
TRAIL < 71.5 pg/mL (n = 97) for the presence of the third tertile of stroke volume was also significantly increased compared to those with serum TRAIL ≥ 71.5 pg/mL (adjusted OR
AC C
EP
[95%CI] was 2.81 [1.61 – 4.92], p < 0.001, Table 4).
ACCEPTED MANUSCRIPT 4. Discussion
In this study, we demonstrated that decreased serum TRAIL levels were associated with the
RI PT
severity of acute ischemic stroke reflected by the NIHSS score and stroke lesion volume on admission. To our knowledge, this is the first report to evaluate the association of serum TRAIL levels with acute ischemic stroke. Several cross sectional and prospective clinical
SC
studies have shown that low serum TRAIL levels are associated with a higher risk for death and worse clinical outcome in patients with various cardiovascular diseases such as chronic
M AN U
heart failure and acute coronary syndrome [6-10]. Our findings suggest that serum TRAIL has important role of apoptosis related brain cell death in acute ischemic stroke. The possible function and role of TRAIL in the cardiovascular system is not completely elucidated. TRAIL leads to apoptotic cell death in tumor cells according to the in vitro studies
TE D
[16,17] and one study reported that soluble TRAIL can kill the tumor cells selectively without apparent toxicity to normal cells in mice [18]. However, there have been conflicting reports for the role of TRAIL in vasculature. Li et al. have shown that soluble TRAIL also induces
EP
apoptosis and pro-inflammatory activity in normal human umbilical vein endothelial cells [19]. On the other hand, several studies have demonstrated possible anti-inflammatory and
AC C
anti-atherogenic activity in vitro and in vivo [4,5,20]. The results of previous clinical studies addressing the relationship between low serum TRAIL and cardiovascular diseases can be explained by the anti-inflammatory property of TRAIL [6-10]. Furthermore, our findings of low serum TRAIL level with the increase of acute ischemic stroke severity also could be additional data, which support the anti-inflammatory role of TRAIL. Osmancik et al. reported that low concentrations of soluble TRAIL represent a strong predictor of a poor prognosis in patients with acute coronary syndrome with 100% ischemic etiology and suggested that TRAIL might be more specific for patients with ischemic
ACCEPTED MANUSCRIPT etiology of left ventricular dysfunction relative to other etiologies [10]. Consistent with a study of Osmancik et al., all patients enrolled our study were also with 100% ischemic etiology of acute stroke regardless of stroke subtype and in agreement with the possible
differences of serum TRAIL levels among stroke subtypes.
RI PT
specificity of TRAIL with ischemic etiology. In addition, there were no significant
The NIHSS is known as a valid, reliable, and reproducible neurologic severity scale [12]
SC
and the NIHSS scores on admission have an association with chronic functional outcome [21] and hospital disposition after stroke [22]. In this study, patients were divided into tertiles
M AN U
according to the NIHSS scores on admission and AUC value for TRAIL that discriminate the presence of highest tertile of NIHSS score was 0.695 [95% CI, 0.617 - 0.773]. We evaluated stroke volume as another parameter for reflecting the severity of acute ischemic stroke. When patients were grouped into the tertiles of stroke volume, AUC value for TRAIL that
TE D
discriminate the presence of highest tertile of stroke volume was somewhat lower than that of NIHSS score (95% CI, 0.634 [0.565 - 0.704]). Volpato et al. have reported that the risk of all cause and cardiovascular disease mortality linearly increases as TRAIL levels decrease near
EP
the value of 70 pg/mL, and after this level there is a curvilinear relationship between TRAIL and the risk of death, with very high mortality risk associated with lowest TRAIL levels in
AC C
1,282 adults aged 65 years or older with pre-existing cardiovascular disease [7]. In our study, the optimal cut-off value of TRAIL for highest tertile of NIHSS score used as a representative of acute ischemic stroke severity was 64.0 pg/mL and 71.5 pg/mL regarding stroke volume, which was quite close to the value of 70 pg/mL in the previous study [7]. Furthermore, Volpato et al. have demonstrated that patients with lowest quartile TRAIL concentrations have a three-fold increased risk of death, compared to those with highest TRAIL levels (Hazard ratios, 3.0; 95% CI, 1.4 - 6.5) among those with prevalent cardiovascular disease [7]. The relative risk of patients with low serum TRAIL for an increase
ACCEPTED MANUSCRIPT of the acute ischemic stroke severity was also significantly higher in our study. This study had some limitations. First, the determination of serum TRAIL levels was based on the measurement of a single sample obtained upon admission and, therefore, we could not
RI PT
evaluate possible changes of serum TRAIL according to the stages of acute ischemic stroke during treatment. Secchiero et al. have shown that serum TRAIL levels were significantly decreased in the early time points after acute myocardial infarction and progressively
SC
increased in the follow-up [23]. With regards to the acute ischemic stroke, reduced serum TRAIL levels at acute stage may be elevated as the recovery of stroke in the follow-up period,
M AN U
which need a further clinical study. Second, the correlation of TRAIL with the acute ischemic stroke severity was weak (r = 0.249 with NIHSS score and r = 0.252 with stroke volume) and the discrimination ability of TRAIL for the ischemic stroke severity was also moderate (AUC, 0.695 using NIHSS score and 0.634 using stroke volume). These results may be related with
TE D
small number of enrolled patients in our study, which limited the statistical power to evaluate the diagnostic value of serum TRAIL for the acute ischemic stroke severity. Finally, the study design was cross-sectional and, as such, causal and longitudinal relationships were not
EP
addressed.
In conclusion, our study has demonstrated that low serum TRAIL levels were associated
AC C
with the severity of acute ischemic stroke for the first time and it does provide additional information regarding relationship between low serum TRAIL level and poor prognosis or outcome of cardiovascular disease. Our findings suggest that serum TRAIL also could be a potential marker for the acute ischemic stroke as well as other cardiovascular diseases.
ACCEPTED MANUSCRIPT Sources of funding This work was supported by Pusan National University Research Grant, 2013.
AC C
EP
TE D
M AN U
SC
Authors had no financial conflicts with commercial entities.
RI PT
Disclosures
ACCEPTED MANUSCRIPT Table 1. Clinical characteristics of patients. Total
Men
Women
Number
293
193
100
Age, years
65.7 ± 11.2
63.8 ± 10.7
69.2 ± 11.3
BMI, kg/m2
23.4 ± 3.0
23.6 ± 2.9
FBG, mg/dL
119.3 ± 28.0
117.8 ± 28.3
Cholesterol, mg/dL
187.2 ± 40.7
187.3 ± 40.5
LDL-C, mg/dL
115.9 ± 36.1
116.2 ± 34.7
115.3 ± 38.6
0.839a
TG, mg/dL
99.0 (73.0 - 142.5)
104.0 (79.0 - 149.0)
85.0 (62.0 - 122.7)
< 0.001b
HDL-C, mg/dL
45.7 ± 13.3
45.0 ± 13.6
45.7 ± 12.7
0.626a
hs-CRP, mg/L
0.15 (0.06 - 0.36)
0.15 (0.07 - 0.39)
0.12 (0.05 - 0.32)
0.178b
Hemoglobin, g/dL
14.4 ± 7.6
15.3 ± 9.2
12.7 ± 1.7
< 0.001a
WBC, ×109/L
7.8 ± 2.9
8.1 ± 3.1
7.5 ± 2.4
0.070a
TRAIL, pg/mL
98.4 (70.1 - 141.0)
100.1 (73.9 - 150.0)
91.5 (64.2 - 129.7)
0.120b
NIHSS score
4 (2 - 7)
4 (2 - 7)
4 (2 - 9)
0.174b
Stroke Volume,
1506.9
1517.0
1477.0
0.845b
(433.7 - 14108.3)
(433.1 - 16144.3)
(433.9 - 12485.1)
191 (65.2%)
120 (62.2%)
71 (71.0%)
0.155c
AC C
Hypertension
P value*
< 0.001a
23.0 ± 3.3
0.157a
121.9 ± 27.3
0.230a
186.9 ± 41.2
0.943a
SC
M AN U
TE D
EP
mm3/L
RI PT
Characteristic
Diabetes mellitus
94 (32.1%)
61 (31.6%)
32 (32.0%)
1.000c
Dyslipidemia
48 (16.4%)
28 (14.5%)
20 (20.0%)
0.246c
Current smoking
140 (47.8%)
132 (68.4%)
8 (0.8%)
< 0.001c
CE 65 (22.2%)
37 (19.2%)
28 (28.0%)
0.103 c
LAA 83 (28.3%)
66 (34.2%)
17 (17.0%)
0.002 c
Stroke subtype
ACCEPTED MANUSCRIPT SVO 95 (32.4%)
60 (31.1%)
35 (35.0%)
0.513 c
OT 50 (17.1%)
30 (15.5%)
20 (20.0%)
0.332 c
Continuous variables are expressed as mean ± standard deviation (SD) or median
RI PT
(interquartile range). The categorical values are given as frequencies and respective percentages.
Abbreviations: BMI, body mass index; FBG, fasting blood glucose; LDL-C, low-density
SC
lipoprotein cholesterol; TG, triglyceride; HDL-C, high-density lipoprotein cholesterol; hsCRP, high-sensitivity C-reactive protein; WBC, white blood cell counts; TRAIL, tumor
M AN U
necrosis factor-related apoptosis inducing ligand; NIHSS, National Institute of Health Stroke Scale; CE, cardioembolism; LAA, large artery atherosclerosis; SVO, small vessel occlusion, OT, stroke of other determined etiology.
Student’s t-test.
b
Mann-Whitney U test.
c
Fisher’s exact test.
AC C
EP
a
TE D
*The P value was calculated by a comparison between men and women.
ACCEPTED MANUSCRIPT Table 2. Relationship between serum TRAIL and risk factors. Parameter
TRAIL, pg/mL
Yes (191)
100.1 (73.1- 438.2)
No (102)
87.8 (19.9 - 346.3)
Yes (94)
99.3 (25.0 - 438.2)
No (199)
96.4 (19.9 - 415.0)
0.281a
M AN U
Dyslipidemia (n) Yes (48)
109.6 (29.4 - 340.0)
No (245)
96.2 (19.9 - 438.2)
0.671a
Yes (140)
98.5 (19.9 - 346.3)
No (153)
TE D
Current smoking (n)
96.5 (25.0 - 438.2)
Stroke subtype (n)
0.045b
85.5 (27.4 - 316.0)
AC C
EP
CE (65)
SVO (95)
SC
0.478a
Diabetes Mellitus (n)
RI PT
0.130a
Hypertension (n)
LAA (83)
P value
OT (50)
102.6 (34.0 - 346.3) 107.7 (34.7 - 438.2) 92.3 (19.9 - 305.0)
Values are expressed as median (interquartile range). Abbreviations: CE, cardioembolism; LAA, large artery atherosclerosis; SVO, small vessel occlusion, OT, stroke of other determined etiology. a
Mann-Whitney U test.
b
Kruskal-Wallis test.
ACCEPTED MANUSCRIPT Table 3. Correlations between log (TRAIL) concentration and various parameters. P value
-0.115
0.065
BMI
0.178
0.004
FBG
-0.132
0.035
Cholesterol
0.152
0.015
LDL-C
0.145
0.020
Log (TG)
0.120
0.054
HDL-C
0.099
0.112
Log (hs-CRP)
-0.102
0.103
Log (NIHSS score)
-0.249
< 0.001
Log (Stroke Volume)
-0.252
< 0.001
M AN U
Age
RI PT
r
SC
Parameters
TE D
Abbreviations: BMI, body mass index; FBG, fasting blood glucose; LDL-C, low-density lipoprotein cholesterol; TG, triglyceride; HDL-C, high-density lipoprotein cholesterol; hs-
AC C
EP
CRP, high-sensitivity C-reactive protein; NIHSS, National Institute of Health Stroke Scale.
ACCEPTED MANUSCRIPT Table 4. Relationships between serum TRAIL and stroke severity. For 3T of NIHSS score
For 3T of stroke volume
TRAIL < 64.0 pg/mL
TRAIL < 71.5 pg/mL
95% CI
P value
ORb
95% CI
P value
7.07
[3.64-13.74] < 0.001
2.81
[1.61-4.92]
< 0.001
RI PT
ORa
Dependent is the presence of the third tertile of NIHSS score or stroke volume.
Odds ratio for patients with TRAIL < 64.0 pg/mL compared to those with TRAIL ≥ 64.0
SC
a
pg/mL.
Odds ratio for patients with TRAIL < 71.5 pg/mL compared to those with TRAIL ≥ 71.5
pg/mL. Adjusted
for
age,
sex,
body
hypercholesterolemia, current smoking.
M AN U
b
mass
index,
hypertension,
diabetic
mellitus,
TE D
Abbreviations: 3T, tertile 3; NIHSS, National Institute of Health Stroke Scale; TRAIL, tumor
AC C
EP
necrosis factor-related apoptosis inducing ligand; OR, odds ratio; CI, confidence interval.
RI PT
ACCEPTED MANUSCRIPT
SC
Fig. 1. Serum TRAIL levels according to the tertiles of NIHSS score (A) and stroke volume (B). Serum TRAIL levels were negatively associated with the severity of acute ischemic
M AN U
stroke.
In each box plot, the lower and upper ends of the box represent the 25th and 75th percentiles (interquartile range, IQR) and the line inside the box represents the median of serum TRAIL (pg/mL). The peripheral lines extend to the outer fences, which are 1.5 times the interquartile
TE D
range below and above the 25th and 75th percentile, respectively.
Abbreviations: NIHSS, National Institute of Health Stroke Scale Score; 1T, tertile 1; 2T, tertile 2; 3T, tertile 3.
EP
Kruskal-Wallis test with Post hoc (Dunn's Multiple Comparison Test) test.
AC C
*, P < 0.05 vs. 1T or 2T.
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
Fig. 2. Receiver-operating characteristic curve for TRAIL to discriminate the severity of acute ischemic stroke. The area under curve (AUC) for the presence of the highest tertile of NIHSS score (A) and the highest tertile of stroke volume (B).
Abbreviations: TRAIL, tumor necrosis factor-related apoptosis inducing ligand; NIHSS,
TE D
National Institute of Health Stroke Scale; AUC, area under the receiver-operating
AC C
EP
characteristic curve; CI, confidence interval.
ACCEPTED MANUSCRIPT References [1]
Manzo F, Nebbioso A, Miceli M, Conte M, De Bellis F, Carafa V, et al. TNF-related apoptosis-inducing ligand: Signalling of a “smart” molecule. The International
[2]
RI PT
Journal of Biochemistry & Cell Biology 41 (2009) 460–466. Wiley SR, Schooley K, Smolak PJ, Din WS, Huang CP, Nicholl JK, et al. Identification and characterization of a new member of the TNF family that induces
[3]
SC
apoptosis. Immunity 3 (1995) 673–682.
Secchiero P, Gonelli A, Carnevale E, Milani D, Pandolfi A, Zella D, et al. TRAIL
M AN U
promotes the survival and proliferation of primary human vascular endothelial cells by activating the Akt and ERK pathways. Circulation 107 (2003) 2250–2256. [4]
Zauli G, Pandolfi A, Gonelli A, Di Pietro R, Guarnieri S, Ciabattoni G, et al. Tumor necrosis factor-related apoptosis-inducing ligand (TRAIL) sequentially upregulates
TE D
nitric oxide and prostanoid production in primary human endothelial cells. Circulation Research 92 (2003) 732–740. [5]
Secchiero P, Candido R, Corallini F, Zacchigna S, Toffoli B, Rimondi E, et al.
EP
Systemic tumor necrosis factor-related apoptosis-inducing ligand delivery shows antiatherosclerotic activity in apolipoprotein E-null diabetic mice. Circulation 114
[6]
AC C
(2006) 1522–1530.
Niessner A, Hohensinner PJ, Rychli K, Neuhold S, Zorn G, Richter B, et al.
Prognostic value of apoptosis markers in advanced heart failure patients. European Heart Journal 30 (2008) 789–796.
[7]
Volpato S, Ferrucci L, Secchiero P, Corallini F, Zuliani G, Fellin R, et al. Association of tumor necrosis factor-related apoptosis-inducing ligand with total and cardiovascular mortality in older adults. Atherosclerosis 215 (2011) 452–458.
[8]
Schoppet M, Sattler AM, Schaefer JR, Hofbauer LC. Osteoprotegerin (OPG) and
ACCEPTED MANUSCRIPT tumor necrosis factor-related apoptosis-inducing ligand (TRAIL) levels in atherosclerosis. Atherosclerosis 184 (2006) 446–447. [9]
Michowitz Y, Goldstein E, Roth A, Afek A, Abashidze A, Ben Gal Y, et al. The
RI PT
involvement of tumor necrosis factor-related apoptosis-inducing ligand (TRAIL) in atherosclerosis. Journal of the American College of Cardiology 45 (2005) 1018–1024. [10]
Osmancik P, Teringova E, Tousek P, Paulu P, Widimsky P. Prognostic Value of
Patients. PLoS ONE 8 (2013) e53860.
Adams HP, Bendixen BH, Kappelle LJ, Biller J, Love BB, Gordon DL, et al.
M AN U
[11]
SC
TNF-Related Apoptosis Inducing Ligand (TRAIL) in Acute Coronary Syndrome
Classification of subtype of acute ischemic stroke. Definitions for use in a multicenter clinical trial. TOAST. Trial of Org 10172 in Acute Stroke Treatment. Stroke 24 (1993) 35–41.
Kasner SE. Clinical interpretation and use of stroke scales. Lancet Neurol 5 (2006)
TE D
[12]
603–612. [13]
Mancia G, De Backer G, Dominiczak A, Cifkova R, Fagard R, Germano G, et al.
EP
2007 Guidelines for the Management of Arterial Hypertension: The Task Force for the Management of Arterial Hypertension of the European Society of Hypertension
AC C
(ESH) and of the European Society of Cardiology (ESC). J Hypertens 25 (2007) 1105–1187.
[14]
American Diabetes Association. Standards of medical care in diabetes--2014.
Diabetes Care 37 (2014) Suppl 1:S14–80.
[15]
Stone NJ, Robinson JG, Lichtenstein AH, Bairey Merz CN, Blum CB, Eckel RH, et al. 2013 ACC/AHA guideline on the treatment of blood cholesterol to reduce atherosclerotic cardiovascular risk in adults: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines. Journal
ACCEPTED MANUSCRIPT of the American College of Cardiology 63 (2014) 2889–2934. [16]
Kischkel FC, Lawrence DA, Chuntharapai A, Schow P, Kim KJ, Ashkenazi A. Apo2L/TRAIL-dependent recruitment of endogenous FADD and caspase-8 to death
[17]
RI PT
receptors 4 and 5. Immunity 12 (2000) 611–620. Sprick MR, Weigand MA, Rieser E, Rauch CT, Juo P, Blenis J, et al. FADD/MORT1 and caspase-8 are recruited to TRAIL receptors 1 and 2 and are
[18]
SC
essential for apoptosis mediated by TRAIL receptor 2. Immunity 12 (2000) 599–609. Walczak H, Miller RE, Ariail K, Gliniak B, Griffith TS, Kubin M, et al. Tumoricidal
5 (1999) 157–163. [19]
M AN U
activity of tumor necrosis factor-related apoptosis-inducing ligand in vivo. Nat Med
Li JH, Kirkiles-Smith NC, McNiff JM, Pober JS. TRAIL induces apoptosis and inflammatory gene expression in human endothelial cells. The Journal of
[20]
TE D
Immunology 171 (2003) 1526–1533.
Secchiero P. TRAIL counteracts the proadhesive activity of inflammatory cytokines in endothelial cells by down-modulating CCL8 and CXCL10 chemokine expression
[21]
EP
and release. Blood 105 (2005) 3413–3419. Frankel MR, Morgenstern LB, Kwiatkowski T, Lu M, Tilley BC, Broderick JP, et al.
AC C
Predicting prognosis after stroke: a placebo group analysis from the National Institute of Neurological Disorders and Stroke rt-PA Stroke Trial. Neurology 55 (2000) 952–959.
[22]
Schlegel DJ, Tanne D, Demchuk AM, Levine SR, Kasner SE, Multicenter rt-PA
Stroke Survey Group. Prediction of hospital disposition after thrombolysis for acute ischemic stroke using the National Institutes of Health Stroke Scale. Arch Neurol 61 (2004) 1061–1064. [23]
Secchiero P, Corallini F, Ceconi C, Parrinello G, Volpato S, Ferrari R, et al. Potential
ACCEPTED MANUSCRIPT Prognostic Significance of Decreased Serum Levels of TRAIL after Acute
AC C
EP
TE D
M AN U
SC
RI PT
Myocardial Infarction. PLoS ONE 4 (2009) e4442.
