See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/315510257

Cholesterol Levels Are Associated with 30-day Mortality from Ischemic Stroke
in Dialysis Patients

Article  in  Journal of stroke and cerebrovascular diseases: the official journal of National Stroke Association · March 2017
DOI: 10.1016/j.jstrokecerebrovasdis.2017.02.007

CITATIONS READS
5 312

272 authors, including:

I-Kuan Wang Tzung-Hai Yen

China Medical University (ROC) Chang Gung Memorial Hospital
177 PUBLICATIONS   1,999 CITATIONS    249 PUBLICATIONS   2,744 CITATIONS   

SEE PROFILE SEE PROFILE

Jiann-Shing Jeng Shih-pin Hsu

National Taiwan University I-Shou University
299 PUBLICATIONS   5,907 CITATIONS    45 PUBLICATIONS   453 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

sleep quality and cognition in mild cognitive impairment and Alzheimer’s disease View project

History-based computerized diagnostic system View project

All content following this page was uploaded by Hao-kuang Wang on 21 February 2018.

The user has requested enhancement of the downloaded file.

 Cholesterol Levels Are Associated with 30-day Mortality from
Ischemic Stroke in Dialysis Patients

I.-Kuan Wang, MD,1,2,3 Chung-Hsiang Liu, MD,4 Tzung-Hai Yen, PhD,5,6

Jiann-Shing Jeng, MD, PhD,7 Shih-Pin Hsu, MD,8 Chih-Hung Chen, MD,9,10
Li-Ming Lien, MD, PhD,11 Ruey-Tay Lin, MD,12 An-Chih Chen, MD,13
Huey-Juan Lin, MD, MPH,14 Hsin-Yi Chi, MD,15 Ta-Chang Lai, MD,16
Yu Sun, MD, PhD,17 Siu-Pak Lee, MD,18 Sheng-Feng Sung, MD,19 Po-Lin Chen, MD,20
Jiunn-Tay Lee, MD,21 Tsuey-Ru Chiang, MD,22 Shinn-Kuang Lin, MD,23
Chih-Hsin Muo, MS,24 Henry Ma, MD,25 Chi-Pang Wen, MD, DrPH,26
Fung-Chang Sung, PhD, MPH,1,24 Chung Y. Hsu, MD, PhD,1,4 and Taiwan Stroke
Registry Investigators

Background: We investigated the impact of serum cholesterol levels on 30-day

mortality after ischemic stroke in dialysis patients. Methods: From the Taiwan
Stroke Registry data, we identified 46,770 ischemic stroke cases, including
1101 dialysis patients and 45,669 nondialysis patients from 2006 to 2013. Results:
Overall, the 30-day mortality was 1.46-fold greater in the dialysis group than in
the nondialysis group (1.75 versus 1.20 per 1000 person-days). The mortality rates
were 1.64, .62, 2.82, and 2.23 per 1000 person-days in dialysis patients with serum

From the 1Institute of Clinical Medical Science, China Medical University, Taichung, Taiwan; 2Department of Internal Medicine, College of
Medicine, China Medical University, Taichung, Taiwan; 3Division of Kidney Disease, China Medical University Hospital, Taichung, Taiwan;
4
Departmemt of Neurology, China Medical University Hospital, Taichung, Taiwan; 5Division of Nephrology, Chang Gung Memorial Hospital,
Taipei, Taiwan; 6Chang Gung University College of Medicine, Taoyuan, Taiwan; 7Neurology, National Taiwan University Hospital, Taipei,
Taiwan; 8Department of Neurology, E-Da Hospital/I-Shou University, Kaohsiung, Taiwan; 9Department of Neurology, College of Medicine,
National Cheng Kung University, Tainan, Taiwan; 10Stroke Center, National Cheng Kung University Hospital, Tainan, Taiwan; 11Department
of Neurology, Shin Kong Wu Ho-Su Memorial Hospital and Taipei Medical University College of Medicine, Taipei, Taiwan; 12Department of
Neurology, Kaohsiung Medical University Hospital, Kaohsiung, Taiwan; 13Department Neurology, Chung Shan Medical University Hospital,
Taichung, Taiwan; 14Department of Neurology, Chi-Mei Medical Center, Tainan, Taiwan; 15Show Chwan Memorial Hospital, Changhua. Taiwan;
16
Cheng Hsin General Hospital, Taipei, Taiwan; 17En Chu Kong Hospital, New Taipei City, Taiwan; 18Department of Neurology, Far Eastern
Memorial Hospital, Taipei, Taiwan; 19Ditmanson Medical Foundation Chia-Yi Christian Hospital, Chia-Yi, Taiwan; 20Taichung Veterans General
Hospital, Taichung, Taiwan; 21Department of Neurology, Tri-Service General Hospital, National Defense Medical Center, Taipei, Taiwan; 22Cathay
General Hospital, Taipei, Taiwan; 23Taipei Tzuchi Hospital, New Taipei City, Taiwan; 24Management Office for Health Data, China Medical
University Hospital, Taichung, Taiwan; 25Monash University, Melbourne, Victoria, Australia; and 26Institute of Population Science, National
Health Research Institute, Zhunan, Taiwan.
Received December 16, 2016; revision received January 19, 2017; accepted February 2, 2017.
Grant support: This study is supported by Taiwan Ministry of Health and Welfare Clinical Trial and Research Center of Excellence (MOHW105-
TDU-B-212-133019), Children’s Hospital of China Medical University (DMR-105-033), China Medical University Hospital (DMR-106-025), Academia
Sinica Taiwan Biobank, Stroke Biosignature Project (BM104010092); NRPB Stroke Clinical Trial Consortium (Most 105-2325-B-039-003); Tseng-
Lien Lin Foundation, Taichung, Taiwan; Taiwan Brain Disease Foundation, Taipei, Taiwan; Katsuzo and Kiyo Aoshima Memorial Funds, Japan.
Address correspondence to Fung-Chang Sung, PhD, MPH, Institute of Clinical Medical Science, China Medical University, 91 Hsueh Shih
Road, Taichung 404, Taiwan. E-mail: fcsung1008@yahoo.com.
1052-3057/$ - see front matter
© 2017 National Stroke Association. Published by Elsevier Inc. All rights reserved.
http://dx.doi.org/10.1016/j.jstrokecerebrovasdis.2017.02.007

Journal of Stroke and Cerebrovascular Diseases, Vol. 26, No. 6 (June), 2017: pp 1349–1356 1349
1350 I.-K. WANG ET AL.
total cholesterol levels of <120 mg/dL, 120-159 mg/dL, 160-199 mg/dL, and
≥200 mg/dL, respectively. Compared to dialysis patients with serum total cho-
lesterol levels of 120-159 mg/dL, the corresponding adjusted hazard ratios of mortality
were 4.20 (95% confidence interval [CI] = 1.01-17.4), 8.06 (95% CI = 2.02-32.2), and
6.89 (95% CI = 1.59-29.8) for those with cholesterol levels of <120 mg/dL, 160-
199 mg/dL, and ≥200 mg/dL, respectively. Conclusions: Dialysis patients with
serum total cholesterol levels of ≥160 mg/dL or <120 mg/dL on admission are at
an elevated hazard of 30-day mortality after ischemic stroke. Key Words:
Cholesterol—dialysis—ischemic stroke, mortality—prognosis.
© 2017 National Stroke Association. Published by Elsevier Inc. All rights reserved.

Introduction to investigate the association between serum total cho-

lesterol levels and 30-day mortality after ischemic
Stroke is one of the leading causes of death and long-
stroke in ESRD patients undergoing dialysis using the
term disability in patients with end-stage renal disease
TSR data.
(ESRD).1 The incidence of stroke is 5-10 times greater in
the ESRD population than in the age-matched general
population.2-5 Factors predisposing ESRD patients to stroke
Materials and Methods
include traditional risk factors for atherosclerotic vascu- Data Source
lar disease, such as age, hypertension, diabetes, and
The TSR program was launched on August 1, 2006 and
dyslipidemia, and nontraditional risk factors associated
prospectively collected information on medical cares for
with a uremic syndrome, such as anemia, abnormal mineral
stroke patients treated at 54 major hospitals in Taiwan,
metabolism, inflammation, malnutrition, and oxidative
with the Institutional Review Board approval.13 To ensure
stress.3
the reliability of the data, neurologists and nurses at the
The risk of death after stroke in long-term dialysis pa-
participating hospitals responsible for completing the reg-
tients is nearly 3 times greater than in the general
istration materials were trained with the TSR’s standard
population.6 Stroke survivors are also at an increased dis-
operating procedure. This study used medical care records
ability risk. 4 Ischemic stroke is more common than
of 83,666 stroke patients who had been documented in
hemorrhagic stroke in ESRD patients.4,5 Although the risk
the TSR database between 2006 and 2013. The registry
of ischemic stroke might increase with high serum cho-
program required each participating hospital to differ-
lesterol levels,7,8 hypercholesterolemia has been associated
entiate types of stroke using computed tomography or
with lower mortality after ischemic stroke in the general
magnetic resonance imaging. All identifications of pa-
population.9-12 The impact of serum cholesterol levels on
tients were scrambled to protect privacy.
mortality after ischemic stroke in dialysis patients has not
been well studied.
Study Subjects
The Taiwan Stroke Registry (TSR) is a government-
funded project that prospectively documented information Figure 1 details the procedure of identifying subjects
on medical cares for acute stroke patients admitted at 54 for the present study. After excluding patients with hem-
major hospitals in Taiwan.13 The aim of this study was orrhagic stroke and other types of stroke, we identified

Figure 1. Flow chart for study subjects.

CHOLESTEROL AND DEATH FROM ISCHEMIC STROKE IN ESRD
60,179 patients with ischemic stroke from 2006 to 2013
through the TSR database. Patients with missing infor-
mation on serum creatinine (N = 4291), serum cholesterol
(N = 5490), or Trial of Org 10172 in Acute Stroke Treat-
ment (TOAST) (N = 3628) classification were excluded.
Among the remaining ischemic stroke patients, 1101 pa-
tients were on dialysis before the stroke and 45,669 patients
were in the nondialysis group.
We extracted information on demographic status, life-
style, subtype of stroke, location of ischemic stroke,
comorbid conditions, blood pressure, laboratory data, Na-
tional Institutes of Health Stroke Scale (NIHSS) at
admission, and lipid-lowering drugs prior to admis-
sion. The ischemic strokes were classified into 5 subtypes:
large-artery atherosclerosis, small-vessel occlusion,
cardioembolism, other determined etiology, and unde-
termined etiology, according to the TOAST criteria.14
Statistical Analysis
Distributions of sex, age, stroke type and location, life-
style, comorbidities, and laboratory data including
cholesterol at admission and NIHSS score at admission
were compared between the dialysis group and the
nondialysis group. The estimated glomerular filtration rate
was calculated by the CKD-EPI (Chronic Kidney Disease
Epidemiology Collaboration) equation for each nondialysis
patient.15 Medians with interquartile ranges were pre-
sented for continuous variables with skewed distributions.
Differences in median values of baseline characteristics
between 2 groups were examined using Wilcoxon rank
sum test. The chi-square test was used to examine the
distributions of categorical variables.
Follow-up for each patient was initiated on the date
of hospital admission until the patient deceased within
30 days or censored due to loss to follow-up, or cen-
sored by 30 days of follow-up. Follow-up person-days
were calculated for each group. The incident 30-day mor-
tality after stroke was calculated, by the serum total
cholesterol levels (<120, 120-159, 160-199, and ≥ 200 mg/dL),
for the dialysis group and nondialysis group separately.
Number of deaths, follow-up person-days, and mortal-
ity were calculated by serum total cholesterol levels. We
used Cox proportional hazards regression analysis to assess
the hazard ratios (HRs) and 95% confidence interval (CI)
of 30-day mortality from ischemic stroke by the serum
total cholesterol levels. Variables with a P value <.25 in
the univariate Cox model were included in the multi-
variate Cox proportional hazards model. Age was forced
in the model. The variable of estimated glomerular fil-
tration rate was also included in the Cox models for
estimating adjusted HRs for the nondialysis group and
both groups combined. The proportional hazards model
assumption was examined using the test of scaled
Schoenfeld residuals. There was no significant relation-
ship between Schoenfeld residuals for serum cholesterol
1351
levels and follow-up time (P = .25). The Cox proportion-
al hazards assumption was valid in this study.
Kaplan–Meier survival curves were plotted and the log-
rank test was used to test the difference. We used SAS
(version 9.3 for Windows; SAS Institute, Cary, NC) to
perform data analyses with a 2-tailed test and a signif-
icance level of P < .05.
Results
Only 2.35% (N = 1101) of the stroke cases were on di-
alysis (Table 1). There were more men than women (60.1%
versus 39.9%) in the study population, with the median
age of 70.0 years. Dialysis and nondialysis groups were
similar in the distributions of sex and age. Compared to
nondialysis patients, the dialysis patients were less likely
to have stroke due to small vessel occlusion (36.9% versus
40%), and had a lower body mass index (median, 23.9
versus 24.3), a lower systolic blood pressure (median,
156 mmHg versus 158 mmHg), a lower hemoglobin
(median, 13.0 g/dL versus 13.8 g/dL), a higher NIHSS
score at admission (11.6% versus 6.61% for >20), a lower
proportion receiving lipid-lowering drug treatment (5.75%
versus 7.42%), and more prevalent history of previous
stroke (30.9% versus 24.1%), ischemic heart disease (17%
versus 12.3%), congestive heart failure (2.48 versus 1.55%),
and diabetes (60.6% versus 39.3%).
Baseline characteristics in dialysis patients with dif-
ferent serum cholesterol levels were shown in Table S1.
There were 372 (33.8%), 260 (23.6%), 224 (20.3%), and
245 (22.3%) patients who had serum total cholesterol levels
of <120, 120-159, 160-199, and ≥ 200 mg/dL, respective-
ly. The gender distribution was similar in the 4 total
cholesterol levels (P = .09). Patients’ age decreased as total
cholesterol increased, but body mass index and hemo-
globin increased with an increase in serum total cholesterol
levels. Patients with serum total cholesterol levels of
160-199 mg/dL had the highest prevalence of atrial fi-
brillation (10.8%), followed by patients with the levels
of 120-159 (6.59%), ≥200 (4.12%), and <120 (1.61%) mg/dL.
The 30-day mortality rate after ischemic stroke was
1.46-fold greater in the dialysis group than in the
nondialysis group (1.75 and 1.20 per 1000 person-days)
(Table 2). Dialysis patients with serum cholesterol levels
of 120-159 mg/dL had the lowest 30-day mortality rate,
.62 per 1000 person-days. Dialysis patients with the
cholesterol level less or greater than this stratum had el-
evated 30-day mortality rates after stroke: 1.64, 2.82,
and 2.23 per 1000 person-days in those with levels
of <120 mg/dL, 160-199 mg/dL, and ≥200 mg/dL, respec-
tively. The Cox model estimated adjusted HRs were 4.20
(95% CI = 1.01-17.4), 8.06 (95% CI = 2.02-32.2), and 6.89
(95% CI = 1.59-29.8), respectively. For nondialysis pa-
tients, the 30-day mortality rate decreased from 1.43 per
1000 person-years in patients with serum total choles-
terol levels of <120 mg/dL to .98 per 1000 person-years
1352 I.-K. WANG ET AL.
Table 1. Baseline characteristics of ischemic stroke patients compared between patients with and without dialysis history, Taiwan
Stroke Registry

All Nondialysis Dialysis

N = 46,770 N = 45,669 N = 1101

Variable n % n % n % P value

Men 28,127 60.1 27,495 60.2 632 57.4 .06*

Age, years .28*
<60 11,990 25.6 11,699 25.6 291 26.4
60-69 11,358 24.3 11,077 24.3 281 25.5
70-79 14,085 30.1 13,751 30.1 334 30.3
≥80 9337 20.0 9142 20.0 195 17.7
Median (Q1, Q3) 70.0 (59.7, 78.2) 70.0 (59.7, 78.2) 69.4 (59.4, 77.4) .16†
TOAST .003*
Large artery atherosclerosis 13,666 29.2 13,359 29.3 307 27.9
Small vessel occlusion 18,661 39.9 18,255 40.0 406 36.9
Cardioembolism 5420 11.6 5292 11.6 128 11.6
Other determined and undetermined etiology 9023 19.3 8763 19.2 260 23.6
Location .17*
No significant finding 13,794 29.5 13,435 29.4 359 32.6
Anterior circulation 18,874 40.4 18,450 40.4 424 38.5
Posterior circulation 9044 19.3 8840 19.4 204 18.5
Both 3227 6.90 3160 6.92 67 6.09
Others 1831 3.91 1784 3.91 47 4.27
BMI, kg/m2 .004
≤20 4442 10.8 4328 10.7 114 11.7
20-24 14,615 35.4 14,229 35.3 386 39.7
>24 22,222 53.8 21,749 54.0 473 48.6
Median (Q1, Q3) 24.3 (22.0, 26.9) 24.3 (22.0-26.9) 23.9 (21.9, 26.8) .03†
Smoking <.0001*
Current 10,244 25.8 10,048 26.0 196 18.5
Past 3910 9.84 3787 9.79 123 11.6
AF 2461 5.29 2404 5.29 57 5.20 .89*
Previous stroke history 11,342 24.3 11,002 24.1 340 30.9 <.0001*
Ischemic heart disease 5778 12.4 5592 12.3 186 17.0 <.0001*
Congestive heart failure 728 1.57 701 1.55 27 2.48 .01*
Diabetes mellitus 16,060 39.8 15,400 39.3 660 60.6 <.0001*
Systolic blood pressure, mmHg .22*
<120 3631 7.77 3527 7.73 104 9.46
120-139 7965 17.0 7766 17.0 199 18.1
140-159 12,436 26.6 12,146 26.6 290 26.4
160-179 10,621 22.7 10,376 22.7 245 22.3
180-199 6748 14.4 6599 14.5 149 13.6
≥200 5329 11.4 5217 11.4 112 10.2
Median (Q1, Q3) 158 (140, 180) 158 (140, 180) 156 (137, 178) .01†
Cholesterol, mg/dL .12*
<120 17,337 37.1 16,965 37.2 372 33.8
120-159 10,157 21.7 9897 21.7 260 23.6
160-199 9424 20.2 9200 20.1 224 20.4
≥200 9852 21.1 9607 21.0 245 22.3
Median (Q1, Q3) 144 (97, 191) 144 (97, 190) 146 (102, 194) .048†
Hb, g/dL <.0001*
<10 2371 5.09 2241 4.93 130 12.1
10-11 6410 13.8 6166 13.6 244 22.6
12-13 16,092 34.6 15,759 34.6 333 30.9
14-15 16,347 35.1 16,071 35.3 276 25.6
≥16 5358 11.5 5263 11.6 95 8.81 <.0001†
Median (Q1, Q3) 13.8 (12.4, 15.0) 13.8 (124, 15.0) 13.0 (11.2, 14.6)
(continued on next page)
CHOLESTEROL AND DEATH FROM ISCHEMIC STROKE IN ESRD 1353

Table 1. (continued)

All Nondialysis Dialysis

N = 46,770 N = 45,669 N = 1101

Variable n % n % n % P value

NIHSS score at admission <.0001*

≤5 31,792 68.0 31,132 68.2 660 60.0
6-20 11,829 25.3 11,516 25.2 313 28.4
>20 3149 6.73 3021 6.61 128 11.6
Median (Q1, Q3) 3 (1, 8) 3 (1, 8) 4 (1, 10) .0003†
eGFR, mL/min/1.73 m2
≥90 8208 18.0
60-89 19,502 42.7
30-59 14,428 31.6
<30 3531 7.73
Lipid-lowering drugs 3469 7.42 3389 7.47 80 5.75 .02*

Abbreviations: AF, atrial fibrillation; BMI, body mass index; eGFR, estimated glomerular filtration rate; Hb, hemoglobin; NIHSS, Nation-
al Institutes of Health Stroke Scale; Q1, 25th percentile; Q3, 75th percentile; SD, standard deviation; TOAST, Trial of Org 10172 in Acute
Stroke Treatment.
*Chi-square test.
†Wilcoxon rank sum test.

Table 2. Hazard ratios of 30-day mortality from ischemic stroke by serum total cholesterol levels at admission compared between
patients with and without undergoing dialysis history

Cholesterol, mg/dL N Death, n Person-days Mortality§ Crude HR (95% CI) Adjusted HR (95% CI)

Dialysis
<120 372 16 9733 1.64 2.64 (.88-7.89) 4.20 (1.01-17.4)*
120-159 260 4 6462 .62 1.00 (reference) 1.00 (reference)
160-199 224 15 5328 2.82 4.50 (1.49-13.5)** 8.06 (2.02-32.2)**
200+ 245 13 5842 2.23 3.58 (1.17-11.0)* 6.89 (1.59-29.8)**
Overall 1101 48 27365 1.75
Nondialysis†
<120 16965 638 445454 1.43 1.27 (1.11-1.47)*** 1.01 (.84-1.20)
120-159 9897 274 239691 1.14 1.00 (reference) 1.00 (reference)
160-199 9200 212 206177 1.03 .88 (.74-1.06) .95 (.76-1.18)
200+ 9607 220 224831 .98 .85 (.71-1.02) 1.00 (.80-1.25)
Overall 45669 1344 1116153 1.20
Overall†‡
<120 17337 654 455187 1.44 1.29 (1.12-1.49)*** 1.04 (.87-1.24)
120-159 10157 278 246153 1.13 1.00 (reference) 1.00 (reference)
160-199 9424 227 211505 1.07 .94 (.78-1.11) 1.02 (.82-1.26)
200+ 9852 233 230673 1.01 .89 (.75-1.06) 1.06 (.85-1.32)
Overall 46770 1392 1143518 1.22

Abbreviations: CI, confidence interval; HR, hazard ratio; eGFR, estimated glomerular filtration rate; NIHSS, National Institutes of Health
Stroke Scale; TOAST, Trial of Org 10172 in Acute Stroke Treatment.
Adjusted for age, gender, TOAST, location, body mass index, smoking, atrial fibrillation, ischemic heart disease, congestive heart failure,
systolic blood pressure, hemoglobin, lipid-lowering drugs, and NIHSS score at admission.
*P < .05; **P < .01; ***P < .001.
†Adjusted further for eGFR levels.
‡Including dialysis and nondialysis patients.
§Per 1000 person-days.

in patients with levels of ≥200 mg/dL. There was no sig- Figure 2 showed the Kaplan–Meier survival curves by
nificant difference in the adjusted hazard of mortality from serum cholesterol levels in both groups. For nondialysis
stroke between patients with serum cholesterol levels of patients, the survival rates increased with serum total cho-
120-159 mg/dL and other subgroup patients. lesterol level, from 95.92% in patients with cholesterol
1354
Figure 2. Survival curves among different serum total cholesterol groups in nondialysis (A) and dialysis patients (B).
of <120 mg/dL to 97.23% in those with cholesterol
of ≥200 mg/dL during the 30-day follow-up (Fig 2, A).
For dialysis patients, patients with serum total choles-
terol levels of 120-159 mg/dL had the highest survival
rate (98.21%), followed by those with levels of <120 mg/dL
(95.14%), ≥200 mg/dL (93.73%) and 160-199 mg/dL (91.98%)
(Fig 2, B).
Discussion
Our study demonstrated that serum total cholesterol
levels of ≥160 mg/dL or <120 mg/dL were associated with
a higher 30-day mortality risk after ischemic stroke in
dialysis patients, compared with those with levels of
120-159 mg/dL. On the other hand, there was no signif-
icant association between serum total cholesterol levels
and the risk of 30-day mortality after stroke in nondialysis
patients after adjustment of age, comorbidities, and stroke
severity.
Previous studies have shown a significant relation-
ship between the ischemic stroke risk and the serum total
cholesterol level for the general population.7,8 Tirschwell
et al found that subjects with the highest total choles-
terol quintile had an odds ratio of 1.6 for ischemic stroke
relative to those with the lowest quintile in a case-
control study.7 On the other hand, higher total serum
cholesterol levels on admission for stroke are associated
with less severe strokes, and better short-term and long-
term outcomes, including deaths.12,16 Vauthey et al reported
that a total serum cholesterol level of greater than
250 mg/dL was an independent predictor of better func-
tional outcomes with a lower risk of severe disability or
death in the first month after acute ischemic stroke.12 Sim-
ilarly, Koton et al found in a prospective registry that the
first-ever ischemic stroke patients with low cholesterol
levels (≤155 mg/dL) have 2-fold greater short- and long-
term mortality rates than those with higher levels,
irrespective of prestroke statin use or not.10 An Italian ret-
rospective study has shown a U-shaped association between
I.-K. WANG ET AL.
serum total cholesterol levels and 30-day mortality in older
patients with acute ischemic stroke. The risk was lower
for patients with normal cholesterol levels (158-200 mg/dL)
than those with higher and lower levels.17 A U-shape re-
lationship between serum cholesterol levels and the 30-
day mortality also appeared among the dialysis patients
in our study. Whereas, the mortality rate in nondialysis
patients decreased as serum cholesterol level increased
in univariate analysis. The risk became insignificant after
controlling for age, comorbidities, and stroke severity.
There are conflicting findings on the relationship between
serum total cholesterol levels and mortality risk in pa-
tients with chronic kidney disease or ESRD.18-21 Studies
have shown that the presence of malnutrition and in-
flammation may modify the relationship between
cholesterol levels and cardiovascular disease.18,20,21 An in-
creased risk of all-cause and cardiovascular mortality
appears in dialysis patients with a higher serum total cho-
lesterol level, but absence of malnutrition and
inflammation.21 On the other hand, dialysis patients with
higher serum total cholesterol levels and presence of mal-
nutrition and inflammation have a lower risk of all-
cause mortality.21 Moreover, several large randomized
clinical trials failed to support a significant efficacy of
cholesterol-lowering therapy with statin in the preven-
tion of cardiovascular mortality for dialysis.22-24
In our study, dialysis status could modify the relation-
ship between cholesterol levels and mortality from ischemic
stroke. We demonstrated a U-shaped relationship between
serum total cholesterol levels and 30-day mortality in-
dependent of stroke severity; patients were at increased
risk if they had serum total cholesterol levels of ≥160 mg/dL
or <120 mg/dL. Our further data analysis showed that,
among the deceased dialysis patients, 10 cases died from
stroke, 2 cases died from heart disease, and the rest died
from complications mostly unspecified. There were fewer
deaths with cholesterol levels of 120-159 mg/dL. The 2
cases who died from heart disease had cholesterol levels
greater than 160 mg/dL.
CHOLESTEROL AND DEATH FROM ISCHEMIC STROKE IN ESRD
Cholesterol is an essential constituent of cell and or-
ganelle membranes, and is critical for the synthesis of
steroid hormone. Low cholesterol levels may reflect mal-
nutrition or inflammation. The FOOD Trial Collaboration
study and a recent study from Taiwan showed that poor
nutritional status on admission was associated with poor
survival and functional status after stroke.25,26 In addi-
tion, cholesterol provides antioxidant protection, limiting
the extent of ischemic injury in animal studies.27 Mice with
hypercholesterolemia have been demonstrated to be more
resistant to lethal hypoxia, and there is inverse associa-
tion between cholesterol levels and survival after exposure
to hypoxia.28 Thus, high cholesterol levels are vascular
risk factor, but low cholesterol levels also contribute to
poor outcome of acute ischemic stroke.
The strength of this study is using a large longitudi-
nal stroke registry data for a follow-up comparison between
dialysis patients and nondialysis patients for the risk of
30-day mortality after ischemic stroke. However, the present
study has several limitations. First, information on the
dialysis modality was not available in the TSR data-
base. Second, laboratory data such as the serum albumin,
C-reactive protein, low-density lipoprotein cholesterol, and
high-density lipoprotein cholesterol were unavailable for
some patients. We were unable to include these vari-
ables in the data analysis for adjustment. Furthermore,
the number of death events was small, thus the statis-
tical power may not be large enough.
Conclusions
The association between serum total cholesterol levels
and the 30-day mortality from ischemic stroke could be
altered by the dialysis status. Ischemic stroke patients on
dialysis with serum total cholesterol levels of ≥160 mg/dL
or <120 mg/dL on admission are at an elevated hazard
for the 30-day mortality.
Appendix: Supplementary Material
Supplementary data to this article can be found online
at doi:10.1016/j.jstrokecerebrovasdis.2017.02.007.
References
1. U.S. Renal Data System. USRDS 2014 annual data report:
atlas of chronic kidney and end-stage renal disease in
the United States. Bethesda, MD: National Institutes of
Health, National Institute of Diabetes and Digestive and
Kidney Diseases, 2014.
2. Kawamura M, Fijimoto S, Hisanaga S, et al. Incidence,
outcome, and risk factors of cerebrovascular events in
patients undergoing maintenance hemodialysis. Am J
Kidney Dis 1998;31:991-996.
3. Seliger SL, Gillen DL, Longstreth WT Jr, et al. Elevated
risk of stroke among patients with end-stage renal disease.
Kidney Int 2003;64:603-609.
1355
4. Sozio SM, Armstrong PA, Coresh J, et al. Cerebrovascular
disease incidence, characteristics, and outcomes in patients
initiating dialysis: the choices for healthy outcomes in
caring for ESRD (CHOICE) study. Am J Kidney Dis
2009;54:468-477.
5. Wang HH, Hung SY, Sung JM, et al. Risk of stroke in
long-term dialysis patients compared with the general
population. Am J Kidney Dis 2014;63:604-611.
6. Herzog CA, Asinger RW, Berger AK, et al. Cardiovascular
disease in chronic kidney disease. A clinical update from
Kidney Disease: Improving Global Outcomes (KDIGO).
Kidney Int 2011;80:572-586.
7. Tirschwell DL, Smith NL, Heckbert SR, et al. Association
of cholesterol with stroke risk varies in stroke subtypes
and patient subgroups. Neurology 2004;63:1868-1875.
8. Zhang X, Patel A, Horibe H, et al. Cholesterol, coronary
heart disease, and stroke in the Asia Pacific region. Int
J Epidemiol 2003;32:563-572.
9. Dyker AG, Weir CJ, Lees KR. Influence of cholesterol
on survival after stroke: retrospective study. BMJ
1997;314:1584-1588.
10. Koton S, Molshatzki N, Bornstein NM, et al. Low
cholesterol, statins and outcomes in patients with first-ever
acute ischemic stroke. Cerebrovasc Dis 2012;34:213-220.
11. Roquer J, Ois A, Rodriguez Campello A, et al. Clustering
of vascular risk factors and in-hospital death after acute
ischemic stroke. J Neurol 2007;254:1636-1641.
12. Vauthey C, de Freitas GR, van Melle G, et al. Better
outcome after stroke with higher serum cholesterol levels.
Neurology 2000;54:1944-1949.
13. Hsieh FI, Lien LM, Chen ST, et al. Get with the
guidelines-stroke performance indicators: surveillance of
stroke care in the Taiwan Stroke Registry: get with the
guidelines-stroke in Taiwan. Circulation 2010;122:1116-1123.
14. Adams HP Jr, Bendixen BH, Kappelle LJ, et al.
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
1993;24:35-41.
15. Levey AS, Stevens LA, Schmid CH, et al. A new equation
to estimate glomerular filtration rate. Ann Intern Med
2009;150:604-612.
16. Olsen TS, Christensen RH, Kammersgaard LP, et al.
Higher total serum cholesterol levels are associated with
less severe strokes and lower all-cause mortality: ten-year
follow-up of ischemic strokes in the Copenhagen Stroke
Study. Stroke 2007;38:2646-2651.
17. Zuliani G, Cherubini A, Atti AR, et al. Low cholesterol
levels are associated with short-term mortality in older
patients with ischemic stroke. J Gerontol A Biol Sci Med
Sci 2004;59:293-297.
18. Contreras G, Hu B, Astor BC, et al. Malnutrition-
inflammation modifies the relationship of cholesterol with
cardiovascular disease. J Am Soc Nephrol 2010;21:2131-
2142.
19. Iseki K, Yamazato M, Tozawa M, et al.
Hypocholesterolemia is a significant predictor of death
in a cohort of chronic hemodialysis patients. Kidney Int
2002;61:1887-1893.
20. Kilpatrick RD, McAllister CJ, Kovesdy CP, et al.
Association between serum lipids and survival in
hemodialysis patients and impact of race. J Am Soc
Nephrol 2007;18:293-303.
21. Liu Y, Coresh J, Eustace JA, et al. Association between
cholesterol level and mortality in dialysis patients: role
of inflammation and malnutrition. JAMA 2004;291:451-
459.
 1356 I.-K. WANG ET AL.
22. Baigent C, Landray MJ, Reith C, et al. The effects of observational data from the FOOD trial. Stroke
lowering LDL cholesterol with simvastatin plus ezetimibe 2003;34:1450-1456.
in patients with chronic kidney disease (Study of Heart 26. Shen HC, Chen HF, Peng LN, et al. Impact of nutritional
and Renal Protection): a randomised placebo-controlled status on long-term functional outcomes of post-acute
trial. Lancet 2011;377:2181-2192. stroke patients in Taiwan. Arch Gerontol Geriatr
23. Fellstrom BC, Jardine AG, Schmieder RE, et al. 2011;53:e149-e152.
Rosuvastatin and cardiovascular events in patients 27. Vatassery GT, Smith WE, Quach HT, et al. In vitro
undergoing hemodialysis. N Engl J Med 2009;360:1395- oxidation of vitamin E, vitamin C, thiols and cholesterol
1407. in rat brain mitochondria incubated with free radicals.
24. Wanner C, Krane V, Marz W, et al. Atorvastatin in Neurochem Int 1995;26:527-535.
patients with type 2 diabetes mellitus undergoing 28. Xi L, Ghosh S, Wang X, et al. Hypercholesterolemia
hemodialysis. N Engl J Med 2005;353:238-248. enhances tolerance to lethal systemic hypoxia in middle-
25. FOOD Trial Collaboration. Poor nutritional status on aged mice: possible role of VEGF downregulation in brain.
admission predicts poor outcomes after stroke: Mol Cell Biochem 2006;291:205-211.

303627 303627 30 36
Chung-Hsiang Liu, 27 31 27 31 27 31 2731 27 31,5727 3127 3127 3127 31 27 32 28 32 2833 28 3328 33
Chon-Haw Tsai, Wei-Shih Huang, Chung-Ta Lu, 29 3329 33 29 33 2934 2934 2934 2934 2934 29 34 29 34 29 3429 3429 35
Tzung-Chang Tsai, Chun-Hung Tseng, Kang-Hsu Lin, 29 35 29 35 29 35 29 35 29 35 2936 2936 2936 29 36 3036
Woei-Cherng Shyn, Yu-Wan Yang, Yen-Liang Liu, Der-Yang Cho, Chun-Chung Chen, Jiann-Shing Jeng, Sung-Chun Tang, Li-Kai Tsai, Shin-Joe Yeh, Shih-Pin Hsu, Han-Jung Chen, Cheng-Sen Chang, Hung-Chang Kuo, Lian-Hui Lee, Huan-Wen Tsui, Jung-Chi Tsou, Yan-Tang Wang, Yi-Cheng Tai, Kun-Chang Tsai, Yen-Wen Chen, Kan Lu, Po-Chao Liliang, Yu-Tun Tsai, Cheng-Loong Liang, Kuo-Wei Wang, Hao-Kuang Wang, Jui-Sheng Chen, Po-Yuan Chen, Cien-Leong Chye, Wei-Jie Tzeng, Pei-Hua Wu, Chih-Hung Chen, Pi-Shan

43 36 50
43 3643 37 43 37 43 37 43 3743 37 43 37,43
44 3850
44 3844 3844 38 4438 4538 45 39 45 40 40 45 4145 4145 42 45 42 45 4245 42 45 4245 53 42
45 43 45 4346 4346 4346 46 43 43 43 43 43 43 43
Sung, Han-Chieh Hsieh, Hui-Chen Su, Hou-Chang Chiu, Li-Ming Lien, Wei-Hung Chen, Chyi-Huey Bai, Tzu-Hsuan Huang, Chi-Ieong Lau, Ya-Ying Wu, Hsu-Ling Yeh, Anna Chang, Ching-Huang Lin, Cheng-Chang Yen, Ruey-Tay Lin, Chun-Hung Chen, Gim-Thean Khor, A-Ching Chao, Hsiu-Fen Lin, Poyin Huang, Huey-Juan Lin, Der-Shin Ke, Chia-Yu Chang, Poh-Shiow Yeh, Kao-Chang Lin, Tain-Junn Cheng, Chih-Ho Chou, Chun-Ming Yang, Hsiu-Chu Shen, An-Chih Chen, Shih-Jei Tsai, Tsong-Ming Lu, Sheng-Ling Kung, Mei-Ju Lee, Hsi-Hsien Chou, Hsin-Yi Chi, Chou-Hsiung Pan, Po-Chi Chan, Min-Hsien Hsu, Wei-Lun Chang,

43 48
Ya-Ying Wu,

48 43 49 43 49
Zhi-Zang Huang,

50 5850
Hai-Ming Shoung,

50 5950 59 50 6050 60 50
Yi-Chen Lo,
44 Fu-Hwa Wang, Ta-Chang Lai, Jiu-Haw Yin, Chung-Jen Wang, Kai-Chen Wang,
46 5546 6356 476356476357 47
Li-Mei Chen,
47 Jong-Chyou Denq, Yu Sun, Chien-Jung Lu, Cheng-Huai Lin, Chieh-Cheng Huang, Chang-Hsiu Liu, Hoi-Fong Chan, Siu-Pak Lee, Ming-Hui Sun, Li-Ying Ke, Po-Lin Chen, Yu-Shan Lee, Sheng-Feng Sung, Cheung-Ter Ong, Chi-Shun Wu, Yung-Chu Hsu, Yu-Hsiang Su, Ling-Chien Hung, Jiunn-Tay Lee, Jiann-Chyun Lin, Yaw-Don Hsu, Giia-Sheun Peng, Chang-Hung Hsu, Chun-Chieh Lin, Che-Hung Yen, Chun-An Cheng, Yueh-Feng Sung, Yuan-Liang Chen, Ming-Tung Lien, Chung-Hsing

60 5168
60 5169
60 5169
60 5169
60 5169
60 527060 527060 52706052 706052706053706053 7060 53716053726053 72 60 53 60 54 61 54 6154 6154 6154 62 63 57 64 57
Chou, Chia-Chen Liu, Fu-Chi Yang, Yi-Chung Wu, An-Chen Tso, Yu-Hua Lai, Chun-I Chiang, Chia-Kuang Tsai, Meng-Ta Liu, Ying-Che Lin, Yu-Chuan Hsu, Tsuey-Ru Chiang, Mei-Ching Lee, Pai-Hao Huang, Sian-King Lie, Pin-Wen Liao, Jen-Tse Chen, Mu-Chien Sun, Tien-Pao Lai, Wei-Liang Chen, Yen-Chun Chen, Ta-Cheng Chen, Wen-Fu Wang, Kwo-Whei Lee, Chen-Shu Chang, Chien-Hsu Lai, Siao-Ya Shih, Chieh-Sen Chuang, Yen-Yu Chen, Chien-Min Chen, Shinn-Kuang Lin, Yu-Chin Su, Cheng-Lun Hsiao, Fu-Yi Yang, Chih-Yang Liu, Han-Lin Chiang, Chun-Yuan Chang, I-sheng Lin, Chung-Hsien Chien, Yang-Chuang Chang, Ping-Kun

6557 6558 6558 6558 6558 65 6559 65 6760 68 60

Chen, Pai-Yi Chiu, Yu-Jen Hsiao, Chen-Wen Fang, Yu-Wei Chen, Kuo-Ying Lee, Yun-Yu Lin, Chen-Hua Li, Hui-Fen Tsai, Chuan-Fa Hsieh, Chih-Dong Yang, Shiumn-Jen Liaw, How-Chin Liao, Shoou-Jeng Yeh, Ling-Li Wu, Liang-Po Hsieh, Yong-Hui Lee, Chung-Wen Chen, Chih-Shan Hsu, Ye-Jian Jhih, Hao-Yu Zhuang, Yan-Hong Pan, Shin-An Shih, Chin-I Chen, Jia-Ying Sung, Hsing-Yu Weng, Hao-Wen Teng, Jing-Er Lee, Chih-Shan Huang, Shu-Ping Chao, Rey-Yue Yuan, Jau- Jiuan Sheu, Jia-Ming Yu, Chun-Sum Ho, Ting-Chun Lin, Shih-Chieh Yu, Jiunn-Rong Chen, Song-Yen Tsai, Cheng-Yu Wei, Chao-Nan Yang, Chao-Hsien Hung,
Ian Shih,
Yip,
Hung-Pin Tseng,
Vinchi Wang, Kaw-Chen Wang, 65
Chin-Hsiung Liu,
65 66 Chun-Liang Lin,
Chung-Fen Tsai,
Hung-Chih Lin,
Chao-Ching Chen,
Pi-Tzu Chen,
Chih-Hao Chen,
Chaur-Jong Hu,
Yi-Chien Liu,
Nai-Fang Chi,
Shao-Yuan Chen,
Lung Chan,
Zi-Hao Zhao,
Chang-Ming Chern,
Zhi-Peng Wei,
Chun-Jen Lin,
Shey-Lin Wu,
Shuu-Jiun Wang,
Ching-Kuan Liu,
Li-Chi Hsu,
Ryh-Huei Lin,
Wen-Jang Wong,
Ching-Hua Chu,
I-Hui Lee,
Sui-Hing Yan,
Der-Jen Yen,
Yi-Chun Lin,
Ching-Piao Tsai,
Pei-Yun Chen,
Shang-Yeong Kwan,
Sheng-Huang Hsiao,
Bing-Wen Soong,
Bak-Sau Yip,
Shih-Pin Chen,
Pei-Chun Tsai,
Kwong-Kum Liao,
Ping-Chen Chou,
Kung-Ping Lin,
Tsam-Ming Kuo,
Chien Chen,
Yi-Chen Lee,
Din-E Shan,
Yi-Pin Chiu,
Jong-Ling Fuh,
Kun-Chang Tsai,
Pei-Ning Wang,
Yi-Sheng Liao,
Yi-Chung Lee,
Ming-Jun Tsai,
Yu-Hsiang Yu,
and Hsin-Yi Kao
Hui-Chi Huang, Jui-Yao Tsai, Ming-Hsiu Wu, Shi-Cheng Chen, Szu-Yi Chiang, Chiung-Yao Wang, Ming-Chin Hsu, Chien-Chung Chen, Po-Yen Yeh, Yu-Tai Tsai, Ko-Yi Wang, Tsang-Shan Chen, Ping-Keung
From the 27China Medical University Hospital; 28National Taiwan University Hospital; 29E-Da Hospital/I-Shou University; 30National Cheng Kung University Hospital; 31Shin Kong WHS Memorial Hospital; 32Kaohsiung Veterans General Hospital; 33Kaohsiung Medical University Chung-Ho Memorial Hospital; 34Chi Mei Medical
Center; 35Chung Shan Medical University Hospital; 36Show Chwan Memorial Hospital; 37Cheng Hsin General Hospital; 38En Chu Kong Hospital; 39Far Eastern Memorial Hospital; 40Kuang Tien General Hospital; 41Taichung Veterans General Hospital; 42Ditmanson Medical Foundation Chia-Yi Christian Hospital; 43Tri-Service General Hospital; 44Cathay General Hospital; 45Changhua Christian Hospital; 46Taipei Tzuchi Hospital; 47Min Sheng General Hospital; 48Lin Shin Hospital; 49National Taiwan University Hospital Yunlin Branch; 50Landseed Hospital; 51Cheng Ching General Hospital; 52China Medical University Beigang Hospital; 53Taipei Medical University -Wan Fang Hospital; 54Taipei Medical University Hospital; 55Kuang Tien General Hospital Dajia Division; 56Changhua Christian Hospital Yunlin Branch; 57Chang Bing Show Chwan Memorial Hospital; 58Lotung Poh Ai Hospital; 59Taipei Medical University—Shuang Ho Hospital; 60Taipei Veterans General Hospital & National Yang-Ming University School of Medicine;

61Chi Mei Medical Center, Liouying; 62Buddhist Dalin Tzu Chi General Hospital; 63St. Martin de Porres Hospital; 64Sin-Lau Hospital, Tainan, the Presbyterian Church in Taiwan; 65Cardinal Tien Hospital; 66Yumin Medical Corporation Yumin Hospital; 67Kaohsiung Municipal Hsiao-kang Hospital; 68Wei Gong Memorial Hospital; 69Taipei City Hospital Ren Ai Branch; 70National Taiwan University Hospital Hsin-Chu Branch; 71Taichung Hospital Department of Health; and 72Tainan Municipal An-Nan Hospital-China Medical University.

View publication stats