Clinical Neurology and Neurosurgery 161 (2017) 98–103

Contents lists available at ScienceDirect

Clinical Neurology and Neurosurgery

journal homepage: www.elsevier.com/locate/clineuro

A randomized controlled study of intravenous fluid in acute ischemic stroke MARK

a,⁎ b c
Nijasri C. Suwanwela , Aurauma Chutinet , Seangduan Mayotarn ,
Ratchayut Thanapiyachaikulb, Napasri Chaisinanunkulb, Thanin Asawavichienjindad,
Sombat Muengtaweepongsae, Yongchai Nilanontf, Jitlada Samajarnb,
Kanokwan Watcharasaksilpg, Somsak Tiamkaoh, Pakkawan Vongvasinkulb, Supparat Charnwutb,
Jeffrey L. Saveri, on behalf of the IVIS Study Group
a
Division of Neurology, Department of Medicine, Chulalongkorn University, Rama IV Road, Bangkok, 10330, Thailand
b
Chulalongkorn Stroke Center, King Chulalongkorn Memorial Hospital, Bangkok, Thailand
c
Division of Neurology, Department of Medicine, Bhumibol Adulyadej Hospital, Bangkok, Thailand
d
Division of Clinical Epidemiology, Department of Medicine, Chulalongkorn University, Bangkok, Thailand
e
Division of Neurology, Department of Medicine, Thammasat University, Pathumtani, Thailand
f
Division of Neurology, Department of Medicine, Siriraj Hospital, Thailand
g
Division of Neurology, Department of Medicine, Chiang mai University, Chiang mai, Thailand
h
Division of Neurology, Department of Medicine, Khon kaen Univerisity, Khon kaen, Thailand
i
Department of Neurology and Comprehensive Stroke Center, David Geffen School of Medicine, University of California, Los Angeles, CA, United States

A R T I C L E I N F O A B S T R A C T

Keywords: Objective: To compare the outcome of patients with acute ischemic stroke who received or did not receive
Ischemic stroke intravenous fluid.
Progressive stroke Patients and methods: This study was a prospective, multicenter, randomized, open-label trial with blinded
Stroke in evolution outcome assessment. We enrolled acute ischemic stroke patients without dehydration aged between 18 and 85
Intravenous fluid
years with NIH Stroke Scale score (NIHSS) score from 1 to 18 who presented within 72 h after onset. Patients
were randomly assigned to receive 0.9% NaCl solution 100 ml/h for 3 days or no intravenous fluid.
Results: On the interim unblinded analysis of the safety data, significant excess early neurological deterioration
was observed among patients in the non-intravenous fluid group. Therefore, the study was prematurely dis-
continued after enrollment of 120 patients, mean age 60 years, 56.6% male. Early neurological deterioration
(increased NIHSS ≥3 over 72 h) not of metabolic or hemorrhagic origin was observed in 15% of the non-IV fluid
group and 3.3% of the IV fluid group (p = 0.02). Predictors of neurological deterioration were higher NIHSS
score, higher plasma glucose, and increased pulse rate. There was no difference in the primary efficacy outcome,
NIHSS ≤ 4 at day 7, 83.3% vs 86.7%, p = 0.61 or secondary efficacy outcomes.
Conclusion: Administration of 0.9% NaCl 100 ml/h for 72 h in patients with acute ischemic stroke is safe and
may be associated with a reduced risk of neurological deterioration. These study findings support the use of
intravenous fluid in acute ischemic stroke patients with NIHSS less than 18 who have no contraindications.

1. Introduction corrected with intravenous normal saline (Class I; Level of Evidence

In patients with acute ischemic stroke, prompt treatment of dehy-
dration is generally recommended, based on the concept that hypovo-
lemia may lead to cerebral hypoperfusion, decreased collateral blood
flow, and aggravation of ischemic brain injury. Conversely, hypervo-
lemia may lead to cerebral edema and cardiac failure. Many guidelines
suggest the assessment of volume status and maintenance of adequate
hydration in acute stroke patients. For example, the American Stroke
Association Guideline 2013 pstates that “Hypovolemia should be
C)”[1]. The European Stroke Organization guideline recommends
normal saline (0.9%) for fluid replacement during the first 24 h after
stroke with a class IV level of evidence [2]. These recommendations are
based on an uncontrolled study demonstrating that higher serum os-
molarity in acute ischemic stroke patients is associated with poor out-
come [3]. However, no randomized controlled studies have compared
the outcome of patients receiving and not receiving intravenous fluid.
We performed a randomized controlled study to compare the clin-
ical outcomes of patients who did and did not receive intravenous fluid.

⁎
Corresponding author.
E-mail address: nijasris@yahoo.com (N.C. Suwanwela).

http://dx.doi.org/10.1016/j.clineuro.2017.08.012
Received 27 June 2017; Received in revised form 14 August 2017; Accepted 23 August 2017
0303-8467/ © 2017 Elsevier B.V. All rights reserved.
N.C. Suwanwela et al.
The primary aim of this study was to evaluate the usefulness and ad-
verse events associated with isotonic solution of 0.9% NaCl adminis-
tered to acute ischemic stroke patients within 3 days after stroke onset.
2. Patients and methods
This study was a prospective, multicenter, randomized, open-label
trial with blinded outcome assessment. The clinical trial was conducted
at 8 centers in different parts of Thailand. The protocol was approved
by the Institutional Review Board at each participating site, and in-
formed consent was obtained from each subject. Data was analyzed by
the Chulalongkorn Research Center, and the safety of the study was
monitored closely by an independent monitoring board. The study was
registered at http://www.controlled-trials.com (unique identifier:
ISRCTN70000879). All patients presenting with acute ischemic stroke
who presented to the participating institutions were screened for study
eligibility between January 2013 and January 2015. The study enrolled
acute ischemic stroke patients aged between 18 and 85 years who
presented at the study site within 72 h after stroke onset and had an
NIH Stroke Scale (NIHSS) score of at least 1 and not more than 18. The
randomization process and treatment were initiated within 24 h after
hospital presentation. We excluded patients who received more than
120 ml of intravenous fluid before randomization; patients who were
candidates for intravenous thrombolysis or required intravenous fluid
for treatment; patients with cardiogenic embolism, history of con-
gestive heart failure or ejection fraction less than 40%; patients with
history of atrial fibrillation or dilated cardiomyopathy; and patients
with renal impairment (serum creatinine > 2 or GFR < 60). All pa-
tients were screened for swallowing dysfunction by a trained nurse
using standard protocol. Those who did not pass the swallowing test,
had tube feeding or receiving any intravenous medications were also
excluded. Evidence of dehydration on clinical examination or urine
specific gravity exceeding 1.030 also precluded patients from the study.
Patients with large infarction in the middle cerebral artery (more than
one-third of its territory) or large cerebellar infarction (more than half
of unilateral cerebellar hemisphere) were excluded. The study patient
flow diagram is shown in Fig. 1.
2.1. Routine patient management
All patients were admitted to an inpatient stroke unit with board-
certified neurologists at each participating centers. They were treated
with standard therapies including antiplatelet agents and general
medical care. Patients were encouraged to consume the same amount of
oral fluid as their previous usual intake.
2.2. Randomization and intervention
Patients were randomly assigned to receive 0.9% NaCl solution (IV
fluid group) or no intravenous fluid (non-IV fluid group) at a ratio of
1:1. Randomization was performed by telephone using a 24-h central
randomization system. For patients allocated to the IV fluid group,
0.9% NaCl solution was administered intravenously at a rate of 100 ml/
h for 72 h. In the non-IV fluid group, no intravenous fluid was given.
Fluid intake/output, urine specific gravity, serum electrolytes, and ad-
verse events, especially neurological deterioration, as well as clinical
symptoms and signs of heart failure were closely monitored during the
treatment period over the first 72 h. Serial NIH Stroke scales were as-
sessed on a daily basis during admission. In case of neurological wor-
sening, a prompt diagnostic evaluation for the cause was undertaken.
Extracranial causes were evaluated by physical examination, vital signs
including oxygen saturation measurements, and laboratory investiga-
tions (complete blood count, serum electrolytes, kidney and liver
function tests). Emergent CT scan of the brain was performed to search
for the causes of neurological deterioration such as brain edema, re-
current stroke, and hemorrhagic transformation. Diagnosis of the cause
99
Clinical Neurology and Neurosurgery 161 (2017) 98–103
of neurological worsening was performed by the treating physician and
also was independently adjudicated by a Clinical Events Committee.
2.3. Follow-up and outcomes
The primary outcome of the study was the proportion of patients
with good outcome defined by NIHSS score equal or less than 4 at day 7
after randomization or at the day of discharge from the hospital,
whichever came first. Secondary outcomes included the proportion of
patients with modified Rankin scale less than or equal to 2 at day 7 after
randomization or the day of discharge, whichever came first, and
proportion of patients with MRS less than or equal to 1 and 2 at 90 days
after randomization. The outcomes were assessed by independent
neurologists, blinded to study arm assignment.
Safety outcomes were defined as the proportion of patients with
heart failure, proportion of patients with neurological deterioration
(worsening of NIHSS by ≥3 points that cannot be explained by meta-
bolic derangements or hemorrhagic transformation of the infarction),
proportion of patients with symptomatic cerebral edema (defined by
neurological worsening together with shifting of the midline structures
on CT scan), and proportion of patients with alteration of serum elec-
trolytes (sodium, potassium, and chloride). Framingham heart failure
risk score was used to screen for congestive heart failure during the
treatment period. The initial safety outcomes were assessed and re-
ported by the investigators and adjudicated by the safety monitoring
committee.
2.4. Statistical analysis
The sample size was calculated based on a previous pilot study in 24
patients (unpublished data), in which good outcome defined by NIHSS
less than 4 at 90 days was found in 75% of the IV fluid group and 60%
of the non-IV fluid group. With α error at 0.05 and ß error 0.10, the
calculated sample size was 215 patients per group. Therefore, to allow
for potential loss to follow-up, we proposed to enroll 500 subjects in
this study. Outcomes were analyzed according to the intention-to-treat
principle, with all randomized patients included in the analysis.
Descriptive statistics were calculated for baseline characteristics for all
subjects by treatment group. For continuous variables, independent t-
tests or Welch t-tests were used to compare the difference in average
value. For discrete variables, Chi-square tests or Fisher exact tests were
used to compare the difference in proportions. The analysis of outcomes
were conducted using logistic regression, adjusted for the following
prespecified prognostic variables: age, sex, body mass index, initial
pulse rate, risk factors (hypertension, diabetes mellitus, dyslipidemia,
smoking, alcohol drinking), initial hematocrit, white blood cell count,
and BUN/Cr ratio. The statistical significance level was defined as a p-
value < 0.05 (two-tailed). Analyses were performed using STATA
software version 11.0 (StataCorp. 2009. Stata Statistical Software: re-
lease 11, College Station, TX: StataCorp LP.) Interim analysis was
planned after enrollment of 100 patients.
Data analysis was performed by Data Management Center of Chula
Clinical Research Center Faculty of Medicine, Chulalongkorn
University.
3. Results
3.1. Study discontinuation
On the interim unblinded analysis of the safety data, excess early
neurological deterioration within 72 h was observed among patients in
the non-IV fluid group. Therefore, the safety monitoring committee
decided to discontinue the study after enrollment of 120 patients.
N.C. Suwanwela et al.
Fig. 1. CONSORT diagram. The number of participants screened, enrolled, completed treatment and analyzed is shown.
3.2. Patients
Of the 120 enrolled patients, 60 each were assigned to the IV fluid
group and non-IV fluid group. Patients’ mean age was 63 years, and
56.6% were male. Patients’ baseline characteristics are shown in
Table 1. The IV fluid group contained more male patients, whereas in
the non-IV fluid group, more smokers and higher mean levels of serum
total cholesterol and high-density lipoprotein were detected.
Total fluid intake was approximately 2.3 times greater in the IV
fluid group compared to the non-IV fluid group. However, oral intake
was not significantly different. Fluid intake in the first 72 h is shown in
Table 2.
3.3. Clinical outcomes
Nine of 60 patients in the non-IV fluid group (15.0%) developed
early neurological deterioration with increased NIHSS score ≥3 during
the first 72 h, whereas 2 of 60 patients (3.3%) in the IV fluid group
deteriorated (p = 0.02). In the non-IV fluid group, deterioration was
first noted a median of 16.1 h (IQR 9.6–44.2) after enrollment and in
the IV fluid group a median of 56.1 h after enrollment (IQR 39.6–72.6).
All patients with deterioration underwent the protocol-defined diag-
nostic evaluation. The independent Clinical Events Committee ad-
judicated the causes of neurologic worsening among non-IV fluid pa-
tients as due to hemorrhagic transformation in 1 patient and due to
possible ischemic stroke progression (no other determined cause) in 8
patients, and among IV-fluid patients as due to ischemic stroke pro-
gression in both of the patients. There were no patients who had sig-
nificant electrolytes imbalance out of the normal range. Following de-
terioration, intravenous fluid was administered to 6 patients in the non-
IV fluid group, according to the judgment of the local investigator.
The NIHSS score on day 7 or day of discharge was comparable
100
Clinical Neurology and Neurosurgery 161 (2017) 98–103
between the 2 groups. The mean NIHSS scores in the IV fluid and non-
IV fluid groups were 2.3 ( ± 3.6) and 2.2( ± 2.8), respectively.
Neurological deterioration and primary and secondary outcomes are
shown in Table 3 (intention-to-treat and per-protocol analyses).
Univariate analysis of factors associated with early neurological
deterioration is shown in Table 4. Besides no IV fluid treatment, female
sex, higher initial NIHSS, and higher pulse rate were associated with
early neurological deterioration. Unadjusted and adjusted odds ratios
for the prediction of early neurological worsening are demonstrated in
Table 5.
Among patients in the non-IV fluid group with neurological dete-
rioration who were administered post-deterioration IV fluid on the
decision of the investigators, 5 had neurological improvement with
NIHSS reduction of more than 2 points within 3 days after IV fluid
administration.
Safety outcomes are shown in Table 6. No patients developed con-
gestive heart failure, and only one patient, in the non-IV fluid group,
had acute myocardial infarction on the third day of treatment.
4. Discussion
Administration of intravenous fluid in patients with acute ischemic
stroke and clear evidence of dehydration is a standard practice, since
hypovolemia has been associated with poor outcome [4]. However,
there is no evidence that intravenous fluid administration is beneficial
for patients without clinical evidence of dehydration.
There are several mechanisms through which intravenous fluid
treatment might benefit acute ischemic stroke patients. First, in-
travenous fluid may expand the intravascular volume, thus improving
cerebral perfusion through collateral circulation. Second, difficulty
swallowing or neurological deficits due to stroke may prevent some
patients from drinking adequate amounts of water, which may lead to
N.C. Suwanwela et al. Clinical Neurology and Neurosurgery 161 (2017) 98–103

Table 1
Baseline characteristics of all patients.

Characteristics IV fluid (n = 60) No IV fluid (n = 60) p value

Age, years (mean ± SD) 60.8 ± 10.0 59.2 ± 10.4 0.41

Male n (%) 40 (66.7) 29 (48.3) 0.04*
Female n (%) 20 (33.3) 31 (51.7)
BMI, kg/m2 (mean ± SD) 24.4 ± 4.1 25.3 ± 4.2 0.27

Risk factor, n (%)

Coronary heart disease 1 (1.7) 3 (5.0) 0.62
Hypertension 35 (58.3) 40 (66.7) 0.35
Diabetes (non-insulin) 9 (15.0) 16 (26.7) 0.12
Dyslipidemia 30 (50.0) 22 (36.7) 0.14
Smoking 32 (53.3) 20 (33.3) 0.03*
Habitual drinking 21 (35.0) 11 (18.3) 0.04*

Clinical presentation, mean ± SD

Onset to treatment, min 1600.7 ± 1069.2 1755.1 ± 1108.6 0.44
Initial NIHSS 3.4 ± 2.2 3.6 ± 2.7 0.77
Systolic BP, mmHg 160.0 ± 27.4 162.3 ± 26.6 0.65
Diastolic BP, mmHg 87.5 ± 16.4 90.7 ± 15.1 0.27
Pulse rate, beats/min 74.3 ± 13.3 78.03 ± 12.7 0.12

Laboratory investigation, mean ± SD

Urine specific gravity 1.013 ± 0.006 1.012 ± 0.016 0.19
Hematocrit, % 42.3 ± 4.8 40.4 ± 4.5 0.03*
WBC*103 8.2 ± 2.2 8.5 ± 2.7 0.59
Plasma glucose, mg/dL 127.7 ± 49.0 146.4 ± 70.0 0.11
HbA1C, % 6.5 ± 2.8 6.1 ± 2.3 0.66
Total cholesterol, mg/dL 201.5 ± 52.0 216.4 ± 64.3 0.46
Triglyceride, mg/dL 150.7 ± 57.8 162.2 ± 79.8 0.64
LDL, mg/dL 131.4 ± 44.6 139.1 ± 60.1 0.68
HDL, mg/dL 38.4 ± 5.5 47.1 ± 13.5 0.01*
BUN, mg/dL 13.5 ± 4.7 13.2 ± 3.8 0.71
Cr, mg/dL 1.0 ± 0.3 0.9 ± 0.2 0.11
BUN/Cr ratio 14.6 ± 4.5 15.8 ± 5.2 0.19

Treatment, n (%)
Antiplatelet in 24 h 60 (100) 60 (100) 1.00
Statin 60 (100) 60 (100) 1.00
Antihypertensive during admission 7(11.67) 4(6.67) 0.34

Table 2 Table 3
Fluid intake during the first 3 days. Neurological deterioration, primary and secondary outcomes of studied patients.

Day Type IV fluid No IV fluid p value Outcome IV fluid n No IV fluid n p-value

(ml) (%) (%)

Day 1 Oral 1098.8 ± 628.9 1296.6 ± 817.5 0.14 Early neurological deterioration
intake Progression of NIHSS ≥ 3 2 (3.3) 9 (15) 0.02
IV fluid 2328.8 ± 471.0 221.4 ± 706.2 < 0.001*
Primary outcome Day 7
Total 3413.3 ± 907.9 1483.0 ± 940.8 < 0.001*
Intention to treat (n = 120)
intake
NIHSS ≤ 4 50 (83.3) 52 (86.7) 0.61
Day 2 Oral 1249.7 ± 553.1 1436.7 ± 670.0 0.10 Per protocol (n = 114)
intake NIHSS ≤ 4 50 (84.8) 49 (89.1) 0.49
IV fluid 2363.0 ± 440.0 286.3 ± 823.9 < 0.001*
Secondary outcome (intention to
Total 3707.1 ± 713.1 1666.6 ± 856.5 < 0.001*
treat)
intake
MRS ≤ 2 day 7 50 (83.3) 52 (86.7) 0.61
Day 3 Oral 1344.98 ± 588.91 1453.90 ± 671.95 0.351 MRS ≤ 1 day 90 50 (83.3) 48 (80.0) 0.64
intake MRS ≤ 2 day 90 53 (88.3) 50 (83.3) 0.43
IV fluid 2356.28 ± 440.57 275.47 ± 796.90 < 0.001*
Other outcome (intention to treat)
Total 3706.35 ± 719.26 1673.00 ± 770.29 < 0.001*
NIHSS ≤ 1 day 7 27 (45.0) 33 (55.0) 0.27
intake
MRS ≤ 1 day 7 44 (73.3) 49 (81.7) 0.27
Mean intake/ 3592.6 ± 578.2 1610.5 ± 738.6 < 0.001*
NIHSS ≤ 1 day 90 42 (70.0) 41 (68.3) 0.84
day
NIHSS ≤ 4 day 90 53 (88.3) 54 (90.0) 0.77

Values shown in mean ± SD.

functions. Large volumes of IV fluid can cause congestive heart failure

relative dehydration. Finally, subclinical dehydration may pre-exist,
and volume overload, which may lead to further complications such as
especially among elderly patients and cases of precipitate acute is-
ischemic heart disease and renal failure. Excess fluid administration
chemic stroke. In an elderly population, plasma osmolarity of acute
may exacerbate brain cytotoxic edema post-stroke.
stroke patients was higher than age-matched healthy controls [5]. Re-
Several methods to improve the hemodynamic status and hemor-
cent study demonstrated that, among patients receiving thrombolytic
rheology in acute ischemic stroke have been proposed. A Cochrane
therapy, dehydration is an independent predictor of poor outcome [6].
systematic review of hemodilution in acute ischemic stroke including
There are also some potential adverse effects of IV fluid adminis-
clinical trials of hydroxyethyl starch, dextran, and albumin showed no
tration, especially among patients with impaired renal and cardiac

101
N.C. Suwanwela et al. Clinical Neurology and Neurosurgery 161 (2017) 98–103

Table 4 clear benefit of these agents on mortality [7]. Moreover, several clinical
Univariate analysis of factors related to early neurological deterioration. trials using hydroxyethyl starch as a volume expander in patients with
acute ischemic stroke were stopped prematurely due to serious com-
Variables No deterioration Early neurological p value
(n = 109) deterioration (n = 11) plications. Major adverse events were coagulopathy with bleeding and
refractory pruritus [8,9]. In this study, we tested isotonic 0.9% NaCl
Age, years 60.0 ± 10.1 59.6 ± 11.1 0.90 solution, since it is widely used and has minimal side effects. In the IV
Male, n (%) 66 (60.6) 3 (27.3) 0.03*
fluid group, isotonic solution (0.9% NaCl) was administered at a rate of
BMI, kg/m2 24.9 ± 4.4 24.6 ± 1.1 0.66
Pulse rate, beats/ 71.6 ± 10.5 80.5 ± 11.2 0.01* 100 ml/h, which is the likely rate of daily fluid balance of healthy adult,
min for a total of 72 h [10,11].
Risk factors, n (%)
This trial aimed to investigate the potential benefit of intravenous
Hypertension 69 (63.3) 6 (54.6) 0.75 fluid treatment in acute stroke patients. However, it was prematurely
Diabetes mellitus 22 (20.2) 3 (27.3) 0.70 terminated due to an excess of neurological deterioration during the
Dyslipidemia 49 (45.0) 3 (27.3) 0.35 first 72-h treatment period after randomization in the non-IV fluid
Smoking 50 (45.9) 2 (18.2) 0.11
group. The potential contribution of non-use of IV fluids to the neuro-
Habitual drinking 31 (28.4) 1 (9.1) 0.28
logic deterioration was supported by the observation that, among pa-
Clinical presentation, mean ± SD
tients allocated to the non-IV group who deteriorated, neurologic def-
Onset to 1685.3 ± 1114.2 1604.4 ± 808.2 0.82
treatment, icits usually improved after administration of 0.9% NaCl.
min Early neurological worsening within 3 days has been found in ap-
Initial NIHSS 2.7 ± 3.2 6.5 ± 3.5 < 0.001* proximately 10–26% of acute ischemic stroke patients [12–14]. Many
score factors have been associated with early neurological deterioration.
Systolic BP, 144.7 ± 24.8 159.9 ± 31.8 0.06
Recent observational studies demonstrated that higher blood urea ni-
mmHg
Diastolic BP, 82.5 ± 15.1 86.4 ± 17.7 0.43 trogen (BUN)/creatinine (Cr) ratio and urine specific gravity were in-
mmHg dependent predictors for clinical worsening [15,16]. These factors may
Laboratory investigation, mean ± SD reflect the relative dehydration of patients. A higher BUN level may also
Urine specific 1.011 ± 0.006 1.013 ± 0.005 0.34 indicate underlying cardiovascular and renal impairment, which could
gravity predispose patients to stroke progression. Our study excluded patients
Hematocrit, % 40.8 ± 5.2 35.0 ± 9.9 0.08 with clinical evidence of dehydration as well as patients who had urine
WBC, cells x 109/L 7.9 ± 2.28 8.2 ± 4.0 0.78
Plasma glucose, 114.5 ± 42.0 113.3 ± 34.8 0.93
specific gravity of more than 1.030. The mean urine specific gravity in
mg/dL our cohort was 0.013, and the mean BUN/Cr ratio was 15. These values
HA1C,% 6.2 ± 1.5 6.1 ± 2.5 0.92 may indicate a minor degree of volume depletion, which is commonly
Total cholesterol, 187.4 ± 51.5 212.0 ± 68.2 0.16 found in acute stroke patients.
mg/dL
Several mechanisms of early neurological deterioration have been
Triglyceride, mg/ 130.4 ± 66.5 155.3 ± 83.1 0.27
dL proposed. Progression of infarction due to collateral insufficiency and
LDL, mg/dL 116.5 ± 45.7 136.5 ± 56.7 0.20 clot propagation are two mechanisms that may be related to hydration
BUN, mg/dL 11.9 ± 3.9 13.1 ± 5.2 0.35 status. Hemodynamic factors including perfusion abnormality have also
Cr, mg/dL 0.9 ± 0.2 0.8 ± 0.2 0.3 been shown to play an important role in patients with early neurolo-
BUN/Cr ratio 13.7 ± 4.1 16.1 ± 6.0 0.09
gical deterioration [17]. Other possible mechanisms might include re-
current embolic stroke, raised intracranial pressure, cerebral edema,
Table 5 and seizure. Improved hydration status with isotonic saline in the IV
Unadjusted and adjusted odds ratio for prediction of early neurological worsening. fluid group may have averted collateral insufficiency or clot propaga-
tion, contributing to the lower proportion of patients with deteriora-
Variable Unadjusted (120) Adjusted (114) tion.
OR (95% CI) OR (95% CI)
This study has limitations. First, since IV fluid administration cannot
be blinded, the study was designed as an open-label trial, which is
Female sex 4.1 (1.03–16.29) 0.6 (0.04–10.23) subject to biases. However, the outcomes were blindly assessed, pre-
Hypertension 0.7 (0.20–2.43) 0.6 (0.07–5.67) specified, and adjudicated by independent experts. Second, although
Diabetes mellitus 1.5 (0.36–6.06) 1.1 (0.13–9.42)
we aimed to investigate patients with NIHSS score < 18, the mean in-
Dyslipidemia 0.5 (0.12–1.82) 0.4 (0.05–3.49)
Smoking 0.3 (0.05–1.27) 0.3 (0.02–4.07) itial NIHSS score in this study was only 3.5. Therefore, the results of this
Habitual drinking 0.3 (0.03–2.05) 0.4 (0.02–8.77) study may be applicable primarily to patients presenting with milder
Pulse rate 1.1 (1.03–1.23) 1.1 (0.98–1.26) neurological deficits. Third, early trial cessation due to group differ-
BMI 1.0 (0.80–1.16) 0.9 (0.71–1.18)
ences in an important safety outcome constrained the study’s power to
Hematocrit 0.9 (0.75–0.96) 0.9 (0.70–1.04)
WBC 1.1 (0.89–1.47) 1.3 (0.71–2.26)
address the primary efficacy hypothesis.
BUN/Cr ratio 1.77 (0.51–6.17) 0.98 (0.12–8.19)

5. Conclusion
Table 6
Adverse events in the study period.
Overall, the results of our study indicate that intravenous fluid is
Adverse events IV fluid, n (%) No IV fluid, n (%)
safe for acute stroke patients, and patients with acute ischemic stroke
who do not receive intravenous fluid are more likely to develop early
Progressive stroke+ 2 (3.3) 9 (15.0) neurological deterioration not of metabolic or hemorrhagic origin. We
Hemorrhagic infarction 0 (0.0) 1 (1.7) conclude that prompt administration of 0.9% NaCl 100 ml/h for 72 h in
GI Bleeding 0 (0.0) 1 (1.7)
Hematuria 0 (0.0) 1 (1.7)
patients with acute ischemic stroke may be associated with a reduced
Seizure 1 (1.7) 0 (0.0) risk of neurological deterioration. These study findings support the use
Myocardial Infarction 0 (0.0) 1 (1.7) of intravenous fluid in acute stroke patients with NIHSS less than 18
who have no contraindications.

102
N.C. Suwanwela et al.
Acknowledgement
The study is funded by the National Research Council of Thailand.
References
[1] E.C. Jauch, J.L. Saver, H.P. Adams Jr, A. Bruno, J.J. Connors, B.M. Demaerschalk,
et al., Guidelines for the early management of patients with acute ischemic stroke: a
guideline for healthcare professionals from the American Heart Association/
American Stroke Association, Stroke 44 (2013) 870–947.
[2] European Stroke Organisation (ESO) Executive Committee, ESO Writing
Committee, Guidelines for management of ischaemic stroke and transient ischaemic
attack 2008, Cerebrovasc. Dis. 25 (2008) 457–507.
[3] A. Bhalla, S. Sankaralingam, R. Dundas, R. Swaminathan, C.D. Wolfe, A.G. Rudd,
Influence of raised plasma osmolality on clinical outcome after acute stroke, Stroke
31 (2000) 2043–2048.
[4] J.W. Schrock, M. Glasenapp, K. Drogell, Elevated blood urea nitrogen/creatinine
ratio is associated with poor outcome in patients with ischemic stroke, Clin. Neurol.
Neurosurg. 114 (2012) 881–884.
[5] G.J. Rodriguez, S.M. Cordina, G. Vazquez, M.F. Suri, J.F. Kirmani, M.A. Ezzeddine,
et al., The hydration influence on the risk of stroke (THIRST) study, Neurocrit. Care
10 (2009) 187–194.
[6] F.F. Wu, Y.C. Hung, Y.H. Tsai, J.T. Yang, T.H. Lee, C.W. Liow, et al., The influence
of dehydration on the prognosis of acute ischemic stroke for patients treated with
tissue plasminogen activator, BMC Cardiovasc. Disord. 17 (June (1)) (2017) 154.
[7] T.S. Chang, M.B. Jensen, Haemodilution for acute ischaemic stroke, Cochrane
103
Clinical Neurology and Neurosurgery 161 (2017) 98–103
Database Syst. Rev. 27 (2014) 8.
[8] C.J. Wiedermann, Complications of hydroxyethyl starch in acute ischemic stroke
and other brain injuries, Pathophysiol. Haemost. Thromb. 33 (2003) 225–228.
[9] R. Woessner, M.T. Grauer, H.J. Dieterich, F. Bepperling, D. Baus, T. Kahles, et al.,
Influence of a long-term, high-dose volume therapy with 6% hydroxyethyl starch
130/0.4 or crystalloid solution on hemodynamics, rheology and hemostasis in pa-
tients with acute ischemic stroke. Results of a randomized, placebo-controlled,
double-blind study, Pathophysiol. Haemost. Thromb. 33 (2003) 121–126.
[10] A. Smorenberg, C. Ince, A.J. Groeneveld, Dose and type of crystalloid fluid therapy
in adult hospitalized patients, Perioper Med. (Lond.) 2 (August (1)) (2013) 17.
[11] M.L. Moritz, J.C. Ayus, Maintenance intravenous fluids in acutely ill patients, N.
Engl. J. Med. 373 (14) (2015) 1350–1360.
[12] J. Roquer, A. Rodríguez-Campello, M. Gomis, J. Jiménez-Conde, E. Cuadrado-
Godia, R. Vivanco, et al., Acute stroke unit care and early neurological deterioration
in ischemic stroke, J. Neurol. 255 (2008) 1012–1017.
[13] P. Birschel, J. Ellul, D. Barer, Progressing stroke: towards an internationally agreed
definition, Cerebrovasc. Dis. 17 (2004) 242–252.
[14] H. Tei, S. Uchiyama, K. Ohara, M. Kobayashi, Y. Uchiyama, M. Fukuzawa,
Deteriorating ischemic stroke in 4 clinical categories classified by the Oxfordshire
Community Stroke Project, Stroke 31 (2000) 2049–2054.
[15] L.C. Lin, W.C. Fann, M.H. Chou, H.W. Chen, Y.C. Su, J.C. Chen, Urine specific
gravity as a predictor of early neurological deterioration in acute ischemic stroke,
Med. Hypotheses 77 (2011) 11–14.
[16] L.C. Lin, J.D. Lee, Y.C. Hung, C.H. Chang, J.T. Yang, BUN/creatinine ratio-based
hydration for preventing stroke-in-evolution after acute ischemic stroke, Am. J.
Emerg. Med. 32 (2014) 709–712.
[17] J.A. Alawneh, R.R. Moustafa, J.C. Baron, Hemodynamic factors and perfusion ab-
normalities in early neurological deterioration, Stroke 40 (2009) 443–450.