ORIGINAL ARTICLE

Nicardipine Infusion for Blood Pressure Control in Patientswith Subarachnoid Hemorrhage

Panayiotis N. Varelas • Tamer Abdelhak •

Jody Wellwood • Irem Shah • Lotfi Hacein-Bey •

Lonni Schultz • Panayiotis Mitsias

Published online: 10 June 2010

� Springer Science+Business Media, LLC 2010

Abstract

Introduction To evaluate the efficacy, tolerability, and

safety of nicardipine infusion in controlling the elevated

blood pressure after subarachnoid hemorrhage (SAH).

Methods Nicardipine infusion was initiated if the indi-

vidual pre-specified systolic blood pressure (SBP) level

goal, mandated by the admitting neurosurgeon, was not

met. Systolic and diastolic BPs were measured on admis-

sion, hourly during the infusion and 12 h before and after

the infusion.

Results Twenty-eight patients with SAH required 50 ni-

cardipine infusions in order to achieve a mean SBP goal of

152 mmHg. The 3,112 extracted BP measurements showed

that mean infusion SBP was significantly lower than

admission and pre-infusion SBP (mean 146.5 vs. 177.1 and

155.6 mmHg, P < 0.001, respectively) and significantly

higher than post-infusion SBP (146.5 vs. 142.6 mmHg,

P = 0.002). Five infusions were stopped prematurely,

because of hypotension (n = 3), emergent surgery (n = 1),

and failure to reach the SBP goal (n = 1). Rebleeding was

not observed in any patient. Nicardipine achieved SBP

control in 59.9% of hourly infusion measurements, with a

trend for higher proportion of success with higher SBP

goals.

Conclusion In this study, nicardipine infusion was a safe

and moderately effective treatment for BP control in

patients with SAH. Although SBP during nicardipine

infusion was higher than the pre-specified goal in a sig-

nificant percentage of hourly observations, this may be due

to the drug administration protocol and other factors such

as analgesia and sedation.

Keywords Nicardipine � Blood pressure � Hypertension �Subarachnoid hemorrhage � Aneurysm

Introduction

Subarachnoid hemorrhage (SAH) is the third most common

cause of stroke (3–5% of all strokes), with an incidence of

10.5/100,000 patient-years [1]. The most common cause of

SAH is rupture of a saccular cerebral aneurysm [2]. These

aneurysms can rebleed, if not secured, a complication that

carries a very high mortality. The risk for rebleeding is

maximum during the first 24 h (4.5–17.3%) [3, 4]. There-

fore, the current management goal is to secure the

aneurysm the earliest possible to avoid this ominous

complication. In the period preceding the surgical clipping

or endovascular coiling of the aneurysm, one of the major

treatment challenges is to control hypertension. This acute

P. N. Varelas (&) � T. Abdelhak � P. Mitsias

Department of Neurology, K-11, Henry Ford Hospital,

2799 West Grand Blvd, Detroit, MI 48202, USA

e-mail: varelas@neuro.hfh.edu

P. N. Varelas � T. Abdelhak � J. Wellwood

Department of Neurosurgery, Henry Ford Hospital, Detroit,

MI, USA

I. Shah

Department of Pharmacy, Henry Ford Hospital, Detroit,

MI, USA

L. Hacein-Bey

Radiological Associates of Sacramento Medical Group Inc,

Sacramento, CA, USA

L. Hacein-Bey

Sutter Neuroscience Institute, Sacramento, CA, USA

L. Schultz

Department of Biostatistics, Henry Ford Hospital, Detroit,

MI, USA

Neurocrit Care (2010) 13:190–198

DOI 10.1007/s12028-010-9393-7

hypertensive response is not uncommon, since in 46.3% of

patients the systolic blood pressure (SBP) can reach 150–

200 mmHg and in 10% of patients >200 mmHg [4]. The

reason for this acute BP elevation after SAH is a surge of

sympathetic output from the brain to the peripheral car-

diovascular system [2].

Nicardipine (Cardene, ESP Pharma, Edison, NJ) is a

dihydropyridine derivative calcium channel blocker, with

potent vasodilatory action. Nicardipine has been used for

blood pressure control in severe hypertension [5], after

ischemic stroke, either intravenously (IV) [6, 7] or intra-

arterially (to improve recanalization of the vessels) [8], to

control elevated BP after intracerebral hemorrhage (ICH)

[9–11] or after traumatic brain injury [11] and to improve

vasospasm after SAH [12–15].

In a recent study, Qureshi et al. reported the feasibility

and safety of treatment of acute hypertension with nicar-

dipine infusion in 29 patients with ICH [9]. Owing to this

experience in ICH and the paucity of reports using nicar-

dipine for controlling acute hypertension after SAH, we

undertook this study. Our aim was to evaluate the efficacy,

tolerability, and safety of nicardipine infusion in control-

ling BP in patients with SAH below an individual pre-

specified level.

Materials and Methods

We prospectively evaluated all patients with non-traumatic

SAH admitted to the Neurosciences Intensive Care Unit

(NICU) at Henry Ford Hospital and treated with nicardi-

pine infusion for BP control in a 37-month period

(December 2005–January 2008). All patients with SAH are

managed by the primary neurosurgical service together

with the neurointensivists (semi-closed ICU model). The

diagnosis of non-traumatic SAH was made by history,

computed tomography of the head, or lumbar puncture. All

patients underwent either four-vessel cerebral angiogram

or computed tomographic angiography to evaluate the

presence of an intracranial aneurysm.

The primary endpoint of the study was the efficacy of IV

nicardipine in controlling SBP below an individual pre-

specified level after SAH. Efficacy was assessed by com-

paring the admission, pre-infusion and post-infusion SBP,

and diastolic BP (DBP) measurements to those during the

infusion and estimating the success rate of keeping the SBP

below the pre-specified level during the infusion. Second-

ary endpoints were tolerability, as assessed by premature

discontinuation of the infusion and safety of this drug

regarding BP control, as assessed by hypotensive events,

other adverse events, and rebleeding rate before discharge

from the hospital. The hospital Institutional Review Board

granted approval of the protocol and data acquisition of this

study.

A specific SBP goal was mandated by the neurosurgical

admitting service for every patient with SAH. As no spe-

cific guidelines for a specific BP goal exist in SAH, this

goal was entirely based on neurosurgeons’ preferences, if

an intracranial pressure monitor was not present. If such a

monitor was placed, the goal was also based on their

preferences, but should at least meet a cerebral perfusion

pressure (CPP) >60 mmHg. The neurocritical care service

managing these patients in the NICU placed an arterial line

in all patients who did not have upon admission to the

NICU. Two groups of patients required nicardipine infu-

sions: those admitted to the NICU on sodium nitroprusside

(SN) infusions and those not responding to the usual IV

antihypertensives administered in the NICU. Per study

protocol, all SN infusions were immediately switched to

nicardipine upon admission to the NICU (transition period

between the two infusions <10 min). In addition, those

patients who failed BP control with IV labetalol (10 mg) or

hydralazine (10 mg) or both (10 min apart) after a waiting

period of 10–20 min were also started on nicardipine. The

nicardipine infusion protocol that we used was proposed by

the manufacturer for gradual BP reduction (http://www.

cardeneiv.com) and was the same used by Qureshi et al.

[9], which prompted us to undertake this study in patients

with SAH. This protocol advocates initiation of therapy at

5 mg/h and increase by 2.5 mg/h every 15 min up to a

maximum of 15 mg/h until the desired BP is achieved. Any

other calcium-channel blocker, including nimodipine, was

held during the nicardipine infusion only. The time to

initiate other antihypertensive medications enterally was

left on the neurointensivist’s discretion. This was usually

done after securing the causative lesion for the SAH and

was followed by weaning the nicardipine infusion off by an

inverse protocol. Short-acting analgesics (fentanyl 25–

100 mcg every 1 h as needed), but no sedatives, were

allowed per protocol because of the need for frequent

neuro-assessments.

All data were entered in a safe database after deletion of

patient identifiers. Patient demographics and other perti-

nent data were collected. The hospital admission systolic

and diastolic BP, as well as hourly BP measurements, up to

12 h before nicardipine infusion, during and up to 12 h

after the infusion of the drug were collected. If >1 ni-

cardipine infusions were required, BP measurements 12 h

before, during, and 12 h after the infusion were also col-

lected. If the time interval between the infusions was less

than 12 h, this period was divided into two equal parts (the

first as a post-infusion interval for the first infusion and the

second as a pre-infusion interval for the second infusion).

As the BP measurements were collected hourly, the time to

Neurocrit Care (2010) 13:190–198 191

achieve the SBP goal was estimated as the number of hours

before the first time that the SBP was below the goal.

Data regarding nicardipine infusions were also collected,

including failure to reach the pre-specified SBP goals during

the infusion, complications and premature discontinuation of

the infusion, adverse events (such as arrhythmias or renal

dysfunction) and concomitant use of other antihypertensive

medications during the weaning period of the infusion. No

patient received nicardipine infusion while vasospasm was

present, but data about vasospasm, if it occurred later during

the course, were also collected. Vasospasm was defined as a

progressive change in mental status not explained by hydro-

cephalus, rebleeding or seizures or a new focal neurologic

deficit, accompanied by abnormal Transcranial Dopplers or

narrowing of the intracranial vessels on vascular neuroim-

aging studies. Vasospasm was classified as mild, for Trans-

cranial Doppler cerebral blood flow velocities of 90–120 cm/s

for the middle cerebral artery, moderate for velocities of

121–200 cm/s and severe for velocities >200 cm/s and a

Lindergaard index >6 [16]. Outcomes included mortality

and modified Rankin Score [17] at hospital discharge.

Statistical Analysis

The mean pre-infusion systolic and diastolic BPs were

compared to the admission blood pressure measurements

using paired t tests. Only the data from the first infusion

were used for this comparison.

The pre-, during, and post-infusion systolic and diastolic

BP measurements were also compared using generalized

estimating equation (GEE) methods. GEE methods are

used to take into account intra- and inter-subject variabil-

ity. In these particular models, patients had multiple BP

measurements recorded during each of the three time

intervals, as well as potentially multiple infusions, all of

which are sources of intra-subject variability. The three

time intervals were considered as repeated measures, and

their pair-wise comparisons were done within each GEE

model using the overall estimates of variability. Random

effects were also assigned for each patient, as well as

multiple infusions within a patient. When assessing dif-

ferences among the treatments that each patient received,

the inter-subject source of variability would be the treat-

ment group. The means and standard errors were also

computed using the GEE methods. The BP measurements

appeared to be normally distributed; hence no additional

data transformations were done. Data from all infusions

were used for these comparisons.

For assessing the relationship between the pre-specified

SBP goal and the proportion of systolic BP measurements

reaching that goal, GEE methods were also used. For each

patient and infusion, the proportion of measurements

meeting the goal was computed across the infusion time

points. This proportion was considered as the outcome of

interest in this GEE model. Random effects were included

for multiple infusions to take into account the intra-subject

variability. The pre-specified SBP goal was included in the

model as a continuous measure and represented the inter-

subject variability. The mean and standard error of the pro-

portions for each SBP goal level were computed taking into

account the possibility of multiple infusions for the same

patient.

As no published data of SBP or DBP absolute changes

with nicardipine infusion exist in the SAH population, we

performed a power analysis after the first ten patients were

enrolled. Using the observed variability estimate for the

difference between admission and infusion SBP, sample

sizes of 154, 40, and 19 patients would have been required

to detect changes of 10, 20, and 30 mmHg with power of

80%, assuming alpha of 0.05 and two-sided testing. Based

on this, we decided to analyze 20–30 patients in the study.

All statistical analyses were performed at the 0.05 level,

using the statistical software SAS version 9.1.

Results

During the study period 184 patients with non-traumatic

SAH were admitted to the NICU. Of those, 156 patients

either had SBP below the pre-specified goals on admission

to the NICU or responded well with the IV labetalol or

hydralazine. Twenty-eight patients were treated with ni-

cardipine infusions during the study period for elevated

SBP above the pre-specified goal, as per the study protocol.

Their demographics, aneurysmal characteristics, treatment,

and outcome are presented in Table 1.

These patients had 50 nicardipine infusions and 3,112

extracted systolic and diastolic BP measurements

(Table 2). Seventeen patients had only one infusion, three

patients two infusions, five patients three infusions, and

three patients four infusions. In one infusion the goal was

changed from 160 to 150 mmHg during the infusion.

SN was infused 17 times before nicardipine. Nicardipine

infusion was started on a median (range) 10.8 (0–582)

hours after admission. During 41/50 nicardipine infusions,

one or more additional per os antihypertensive medications

were initiated (calcium channel blockers in three infusions,

in violation of the protocol).

Five infusions had to be stopped prematurely because

of hypotension in three patients (after receiving fentanyl,

midazolam, hydralazine, and labetalol IV in one patient,

metoprolol, furosemide IV in another, >1 enteral anti-

hypertensive medications in the third), emergent surgery

in one patient, and failure to reach the SBP goal after

several hours in one (SN infusion was started instead).

192 Neurocrit Care (2010) 13:190–198

No patient developed rebleeding until discharge, renal

dysfunction, or arrhythmias during the infusion.

Four data collection periods for the first infusion of each

patient (admission, pre-infusion, infusion, and post-infu-

sion) and three for each additional infusion (pre-infusion,

Table 1 Demographics and other descriptive characteristics of the

patients

Variable Patients

(n = 28)

Age (mean ± SD) 59.0 ± 14.3

Male sex, n (%) 10 (36)

Race, n (%)

Caucasian 9 (32)

African American 19 (68)

History of hypertension, n (%)

Yes 20 (71)

No 6 (21)

Unknown 2 (8)

Glasgow Coma Scale on admission (mean ± SD) 11.9 ± 3.4

Hunt and Hess grade, n (%)

I 5 (18)

II 7 (25)

III 9 (32)

IV 6 (21)

V 1 (4)

Fisher grade, n (%)

1 –

2 2 (8)

3 13 (46)

4 13 (46)

Intraventricular blood, n (%) 18 (64)

External ventricular drainage placement, n (%) 19 (68)

Location of aneurysm or lesion, n (%)

Internal carotid artery 6 (21)

Middle cerebral artery 1 (3.5)

Anterior cerebral artery/pericallosal artery 5 (18)

Anterior communicating artery 5 (18)

Posterior communicating artery 3 (11)

Basilar artery 2 (7)

Foramen magnum arteriovenous malformation/

fistula

1 (3.5)

No lesion found 5 (18)

Treatment of aneurysm, n (%)

Clipping 8 (29)

Coiling 11 (39)

Aneurysm found—no treatment 4 (14)

No aneurysm found—no treatment 5 (18)

Vasospasm, n (%) 14 (50)

Vasospasm severity, n (%)

Mild 3 (21)

Moderate 6 (42)

Severe 5 (37)

Shunt placed, n (%) 6 (21)

mRankin at discharge, n (%)

0 1 (4)

1 2 (7)

Table 2 Descriptive information for the nicardipine infusions

Variable Infusions

(N = 50)

Systolic BP on admission (mean ± SD) 177.1 ± 41.5

Diastolic BP on admission (mean ± SD) 86.4 ± 22.7

Systolic BP pre-specified goal (mean ± SD) 152 ± 11.2

Infusions with pre-specified SBP goal, n (%)

<130 mmHg 1 (2)

<140 mmHg 14 (28)

<150 mmHg 14 (28)

<160 mmHg 17 (34)

<180 mm Hg 3 (6)

Time interval between admission and first

nicardipine infusion (min, mean ± SD)

66.2 ± 134.8

SN infusion before nicardipine, n (%) 17 (34)

Nicardipine maximum dose, n (%)

B5 mg/h 20 (40)

6–10 mg/h 20 (40)

>10 mg/h 10 (20)

Nicardipine maximum dose (mean ± SD) 8.2 ± 3.8

Nicardipine duration in hours (mean ± SD) 14.2 ± 18.2

Time to achieve the goal in hours (mean ± SD) 3 ± 4.1a

Premature discontinuation of infusion, n (%) 5 (10)b

Concomitant use of other anti-hypertensives, n (%)

Labetalol or b-blocker 5 (11)

Hydralazine 3 (6)

Nimodipine or CCB 3 (6)

More than one antihypertensive 30 (64)

No other anti-hypertensive 6 (13)

BP blood pressure, SN sodium nitroprusside, CCB calcium-channel

blockera One patient never achieved the goalb Three for hypotension, one during surgery, and one for failure to

reach systolic BP goal

Table 1 continued

Variable Patients

(n = 28)

2 2 (7)

3 2 (7)

4 13 (48)

5 4 (15)

6 3 (11)

Neurocrit Care (2010) 13:190–198 193

infusion and post-infusion) were compared. The mean

infusion SBP was significantly lower than the admission and

pre-infusion SBP (mean 146.5 vs. 177.1 and 155.6 mmHg,

P < 0.001, respectively) and significantly higher than the

post-infusion SBP measurements (146.5 vs. 142.6 mmHg,

P = 0.002, Fig. 1). The mean infusion DBP was also sig-

nificantly lower than the admission and pre-infusion DBP

(mean 69.8 vs. 86.4 and 75.9 mmHg, P < 0.001, respec-

tively), and not different than the post-infusion DBP

measurements (69.8 vs. 69.6 mmHg, P = 0.8, Fig. 2).

On the average, nicardipine achieved control at the

pre-specified SBP in 59.9% (SE 5.6%) of all hourly SBP

measurements during infusions, with a trend for higher

proportion of success with higher SBP pre-specified goals

(Fig. 3). In order to elucidate why the neurosurgeons

allowed such an SBP goal variability, we examined the

relationship of the pre-specified SBP goal with factors

such as the age of the patient or the severity of the

clinical (Glasgow Coma Scale, Hunt & Hess grade) and

radiographic (Fisher grade) presentation, but we did not

find any significant correlation (Pearson coefficient

>0.05). The mean time to achieve an SBP below the goal

was 3 h (median 1.3 h, range 0–15 h), with one patient

never achieving the goal after 15 h of infusion. Of note,

this patient received nicardipine at max dose 10 mg/h

during this unsuccessful infusion per violation of the

protocol and had three other nicardipine infusions with

the same SBP goal, which controlled the BP in 12, 2, and

0 h.

We also performed a stratification analysis comparing

the four BP data collection periods in specific subgroups

of patients: those with and without SN, those receiving no

additional, one additional, and >1 additional antihyper-

tensive medications during the weaning phase of the

infusions and those receiving low-, medium-, and high-

dose infusions (B5, 6–10, and 11–15 mg/h, Table 3). We

also compared the average BP measurements of each of

the four BP data collection periods between the subgroups

(not presented in Table 3): there was a significant differ-

ence between the SN subgroups regarding the admission

and pre-infusion mean SBP and DBP (admission SBP

P < 0.001, pre-infusion SBP P = 0.004, admission DBP

P < 0.001, and pre-infusion DBP P = 0.035). There was

also a significant difference between the SN subgroups for

DBP at post infusion (P = 0.042). In the subgroups with

the additional antihypertensives, the admission SBP

and DBP were significantly higher in the subgroup with

one antihypertensive than the other subgroups (no anti-

hypertensive SBP P < 0.001, >1 antihypertensive SBP

P = 0.004, no antihypertensive DBP P < 0.001, and >1

antihypertensive DBP P = 0.027). In addition, the dif-

ferences between the subgroup with no antihypertensive

and the other two groups were significant for pre-infusion

SBP (P = 0.019 for both comparisons). There was

no difference in mean BP between the three infusion

doses regarding any of the periods, except high- and

medium-dose SBP (P = 0.001) and high- and low- or

medium-dose DBP (P < 0.001, both) for the post-infu-

sion period.

130

140

150

160

170

180

190S

BP

in m

m H

g

Admission Pre-Infusion During Infusion Post Infusion

Fig. 1 Mean SBP measurements during the four periods for all

infusions. Vertical bars represent ± SE. For all pair-wise compari-

sons the P-values were <0.01

60

70

80

90

100

Admission Pre-Infusion During Infusion Post Infusion

DB

P in

mm

Hg

Fig. 2 Mean diastolic blood pressure measurements during the four

periods for all infusions. Vertical bars represent ± SE. For all pair-

wise comparisons the P-values were <0.05, except for during

infusion versus post-infusion (P = 0.8)

0

10

20

30

40

50

60

70

80

140 150 160 180

Systolic BP Goal in mm Hg% o

f S

ucc

ess

to m

eet

the

SB

P p

re-s

et g

oal

Fig. 3 Success rate for controlling the SBP based on the pre-

specified SBP goal

194 Neurocrit Care (2010) 13:190–198

Ta

ble

3P

air-

wis

eco

mp

aris

on

of

the

thre

ep

erio

ds,

pre

-in

fusi

on

,in

fusi

on

,an

dp

ost

-in

fusi

on

inth

eS

Nsu

bg

rou

ps,

the

sub

gro

up

sw

ho

rece

ived

add

itio

nal

anti

hy

per

ten

siv

em

edic

atio

ns

du

rin

gth

en

icar

dip

ine

infu

sio

nan

dth

esu

bg

rou

ps

wit

hlo

w-

(B5

mg

/h),

med

ium

-(6

–1

0m

g/

h),

and

hig

h-d

ose

(11

–1

5m

g/h

)n

icar

dip

ine

infu

sio

ns

No

SN

SN

No

AH

M1

AH

M>

1A

HM

Lo

w-d

ose

infu

sio

n

Med

ium

-do

se

infu

sio

n

Hig

h-d

ose

infu

sio

n

SB

Pm

ean

(SE

)

Ad

mis

sio

n1

52

.8(6

.2)

19

8.8

(5.7

)1

62

.6(8

.9)

20

7.6

(9.6

)1

72

.0(6

.1)

13

9.3

(13

.9)

17

8.3

(5.4

)1

78

.5(8

.2)

Pre

-in

fusi

on

15

0.3

(3.3

)1

62

.6(3

.2)

14

9.4

(5.9

)1

62

.5(4

.1)

15

7.5

(3.7

)1

48

.4(3

.3)

15

5.8

(3.3

)1

57

.2(4

.1)

Du

rin

gin

fusi

on

14

5.6

(3.2

)1

46

.9(2

.9)

14

6.4

(5.9

)1

41

.4(2

.8)

14

8.3

(3.6

)1

47

.5(2

.9)

14

6.3

(3.2

)1

49

.0(3

.7)

Po

stin

fusi

on

13

9.7

(3.2

)1

45

.8(3

.2)

14

1.5

(5.8

)1

40

.2(2

.8)

14

4.0

(3.7

)1

44

.2(2

.2)

13

7.5

(3.2

)1

55

.6(3

.9)

P-v

alu

esfo

rp

air-

wis

eco

mp

aris

on

s

Ad

mis

sio

nv

ersu

sp

re-i

nfu

sio

n0

.65

3<

0.0

01

0.0

76

<0

.00

10

.00

60

.52

5<

0.0

01

0.0

06

Ad

mis

sio

nv

ersu

sd

uri

ng

0.1

96

<0

.00

10

.03

0<

0.0

01

<0

.00

10

.56

7<

0.0

01

<0

.00

1

Ad

mis

sio

nv

ersu

sp

ost

-in

fusi

on

0.0

18

<0

.00

10

.00

4<

0.0

01

<0

.00

10

.72

5<

0.0

01

0.0

03

Pre

-ver

sus

du

rin

gin

fusi

on

0.0

12

<0

.00

10

.43

0<

0.0

01

<0

.00

10

.84

5<

0.0

01

0.0

01

Pre

-ver

sus

po

st-i

nfu

sio

n<

0.0

01

<0

.00

10

.01

6<

0.0

01

<0

.00

10

.32

1<

0.0

01

0.5

78

Du

rin

gv

ersu

sp

ost

-in

fusi

on

<0

.00

10

.58

70

.15

10

.69

00

.00

80

.41

0<

0.0

01

0.0

06

DB

Pm

ean

(SE

)

Ad

mis

sio

n7

5.9

(3.7

)9

4.1

(4.4

)7

2.1

(6.3

)9

8.5

(7.0

)8

3.2

(3.6

)6

6.8

(8.6

)8

6.8

(3.3

)8

6.9

(4.6

)

Pre

-in

fusi

on

73

.4(2

.1)

79

.0(3

.5)

73

.6(3

.9)

81

.2(4

.6)

74

.2(2

.3)

65

.3(2

.5)

75

.3(2

.2)

81

.4(2

.2)

Du

rin

gin

fusi

on

68

.7(2

.0)

71

.0(3

.4)

65

.2(4

.0)

70

.0(4

.3)

69

.9(2

.2)

57

.4(2

.3)

70

.7(2

.2)

73

.3(1

.9)

Po

stin

fusi

on

67

.3(2

.0)

72

.8(3

.5)

67

.0(3

.9)

68

.1(4

.4)

70

.2(2

.2)

55

.4(2

.1)

67

.8(2

.2)

79

.4(2

.1)

P-v

alu

esfo

rp

air-

wis

eco

mp

aris

on

s

Ad

mis

sio

nv

ersu

sp

re-i

nfu

sio

n0

.45

6<

0.0

01

0.7

86

0.0

04

0.0

02

0.8

69

<0

.00

10

.22

6

Ad

mis

sio

nv

ersu

sd

uri

ng

0.0

31

<0

.00

10

.20

4<

0.0

01

<0

.00

10

.29

2<

0.0

01

0.0

02

Ad

mis

sio

nv

ersu

sp

ost

-in

fusi

on

0.0

10

<0

.00

10

.34

8<

0.0

01

<0

.00

10

.18

6<

0.0

01

0.0

90

Pre

-ver

sus

du

rin

gin

fusi

on

<0

.00

1<

0.0

01

<0

.00

1<

0.0

01

<0

.00

10

.00

6<

0.0

01

<0

.00

1

Pre

-ver

sus

po

st-i

nfu

sio

n<

0.0

01

<0

.00

10

.00

3<

0.0

01

<0

.00

1<

0.0

01

<0

.00

10

.22

5

Du

rin

gv

ersu

sp

ost

-in

fusi

on

0.1

59

0.1

38

0.4

54

0.2

77

0.7

60

0.4

44

0.0

02

<0

.00

1

SB

Psy

sto

lic

blo

od

pre

ssu

re,

DB

Pd

iast

oli

cb

loo

dp

ress

ure

,S

Nso

diu

mn

itro

pru

ssid

e,A

HM

anti

-hy

per

ten

siv

em

edic

atio

n

Neurocrit Care (2010) 13:190–198 195

Discussion

This study suggests that nicardipine is an effective medi-

cation for BP control in patients with SAH. Both the mean

SBP and DBP were significantly lower during the infusion

period compared to the admission and pre-infusion periods

(Fig. 1). Although additional factors may have contributed

to this difference, since there was already a decline in the

BP from admission to pre-infusion, this decline may lar-

gely be explained by the use of SN (Table 3, no significant

difference between admission versus pre-infusion in the

subgroup without SN) or other IV antihypertensives, such

as labetalol or hydralazine (also no significant difference in

the subgroup without additional antihypertensives). The

effect of the nicardipine infusion may also have extended

to the post-infusion period, allowing the BP to continue

declining for several reasons, including short post-infusion

recording periods with residual drug effect (especially at

medium-dose infusions) or, most likely the addition of

other enteral antihypertensives. Indicative of the latter is

the presence of a difference between infusion and post-

infusion SBP in the subgroup with >1 antihypertensives

(administered in an attempt to wean the nicardipine toward

the end of the infusion time).

Previous studies with IV nicardipine use in patients with

SAH reported use for prolonged periods (up to 14 days

from onset) and with endpoints not blood pressure control:

all these studies have been focusing on dose finding, tol-

erability, vasospasm development, and outcomes [12–15].

Only recently Roitberg et al. examined the safety and

efficacy of nicardipine versus SN in patients with either

SAH or ICH. Although it is not clear how many treated

patients belonged to each subgroup of hemorrhagic stroke,

the goal of this study, similar to ours, was utilization of

these infusions for BP control below pre-specified levels

[18]. In another recent study, Liu-Deryke et al. compared

the use of nicardipine or labetalol in patients with stroke.

Only six patients with SAH were treated with nicardipine

in this retrospective study [19]. Lastly, in a recent study

with a mixed population, including diagnoses of traumatic

brain injury, SAH, ICH, arteriovenous malformation, and

hypoxic brain injury, Narotam et al. reported that nicardi-

pine was effective in reducing the mean arterial BP by

19.7% after 4 h and did not affect the regional brain tissue

oxygen, despite a reduction in CPP. Similar to the study by

Roitberg et al., this study does not report on the specific

effects of the drug in the subgroup with SAH (11 patients)

[11]. Therefore, our study is the first that exclusively

evaluated the intravenous use of this dihydropyridine

exclusively for BP control in a large number of patients

with SAH. It is worth mentioning that IV nicardipine has

already been included in the published guidelines for BP

control in ischemic stroke [6] and ICH [20]. The recently

published guidelines for SAH [21] also include nicardipine,

but, surprisingly, without available data for such a use.

Nicardipine failed to reach or maintain the SBP below

the pre-specified goal in 40.1% of the hourly measure-

ments. This medication, however, was deemed inadequate

for BP control and had to be changed to SN in only one

infusion (2%). Our results mirror those by Roitberg et al.,

who found BP to be outside the pre-specified goal in 38%

of the nicardipine infusion time [18] and Liu-Deryke et al.,

who found that 40% of nicardipine-treated patients with

ICH failed to meet the BP goal within 24 h [19]. There are

several potential contributors to this suboptimal BP

response. First, patients with SAH usually have severe

headache, responding primarily to opioids. Many neuro-

intensivists, including us, prefer to use short-acting opioids

in low, frequent doses to avoid over-sedation. As a con-

sequence, frequent rebound hypertension and headache or

agitation ensue, making titration of anti-hypertensives a

difficult task. Second, cerebral edema and hydrocephalus

are also frequent complications of SAH leading to acute

hypertension in order to maintain adequate cerebral per-

fusion pressure. After a ventriculostomy is placed and

cerebrospinal fluid is drained, the elevated BP may decline

requiring close titration of the antihypertensive. Third, the

infusion protocol that we used may have not been

aggressive enough to allow faster and tighter titration.

Although the same protocol was used in patients with ICH

and led to mean arterial BP control in 86% of the time [9],

these patients are not similar to ours and mean BP may

have less variability than SBP, allowing easier titration of

the infusion. A protocol with adjustment of the rate every

five instead of 15 min is supported by the manufacturer

now and might have allowed a lower failure rate. In fact,

compared to SN, nicardipine infusion with 5-min adjust-

ments leads to more rapid control of post-operative

hypertension and fewer dose adjustments [22].

Lastly, we found that the lower the pre-specified SBP

goal was, the higher the failure of nicardipine infusion

(Fig. 3), suggesting that a more ‘‘relaxed’’ BP goal might

have led to better BP control with the nicardipine infusions.

Although lower BP may theoretically decrease rebleeding,

this has not been demonstrated sufficiently [23]. In a ret-

rospective study, systolic BP C160 mmHg was a risk

factor of rebleeding, while SBP B140 was not [24]. In

another observational study, rebleeding occurred less fre-

quently in patients with antihypertensive treatment (15%)

compared to patients without anti-hypertensive treatment

(33%). The group with anti-hypertensive treatment had,

however, more frequent cerebral infarctions [25]. There-

fore, the optimal blood pressure level after SAH may be

dependent on multiple variables and be unknown in the

individual patient. In fact, rebleeding may be due to

the variations or changes in blood pressure rather than the

196 Neurocrit Care (2010) 13:190–198

absolute BP level [26]. Echoing this, the most recent

guidelines for the management of SAH recommend that

‘‘BP should be monitored and controlled to balance the risk

of stroke, hypertension-related rebleeding, and mainte-

nance of cerebral perfusion pressure’’ as Class I, Level of

Evidence B [21]. Until better data and more specific

treatment algorithms are developed, the optimal BP level

may still vary considerably and be dependent on the

admitting physician’s preferences.

Nicardipine was also a safe medication to use. No reb-

leeding occurred in these patients, which may be due to a

combination of BP control and early treatment of the culprit

lesion. Vasospasm occurred in 50% of patients, identical to

that found in the study, which compared nicardipine to SN

[18]. The three deaths that occurred in our patient population

were temporally unrelated to nicardipine. Only five infusions

(10%) were prematurely discontinued, mostly because of

hypotension induced by additional sedatives, analgesics, or

anti-hypertensives in combination with nicardipine. Previ-

ous studies have reported 3.3–28% incidence of hypotension

with 14-day nicardipine infusions [14, 15]. We believe that

these medications should be carefully administered in these

patients, starting with one drug, at low dose, and escalating

under close BP monitoring.

Our study has limitations. Although all patients had non-

traumatic SAH, the etiology of the bleeding was non-

aneurysmal in 3.5% and no lesion was found in another

18% of patients. The natural history of aneurysmal may be

different than non-aneurysmal SAH, making thus our study

population not very homogeneous. The pre-specified SBP

goals were chosen by the neurosurgeons based on personal

preferences for the individual patient. It is unclear which

factors influenced each neurosurgeon to adopt a specific

SBP goal, because age or clinical or radiographic severity

at presentation did not correlate with these goals. There

was no range of SBP, as the goal was to treat elevated BP.

The lower acceptable SBP limit was the same for every

patient, as hypotension is defined in our unit as SBP

<100 mmHg in the admission orders. A larger number of

patients could have given a better assessment of the risk for

rebleeding even under optimal BP control. There was no

control arm in the study. We believed that it would be

unethical to use SN as comparator drug (since it may

increase the intracranial pressure and is rarely used in

NICUs, especially in patients with SAH, where the ICP

may be already elevated from hydrocephalus), or placebo,

in a patient population with the potential lethal complica-

tion of rebleeding from acute hypertension. The role that

the additional anti-hypertensives played in controlling the

BP and allowing weaning the infusion off is not clear.

Except for the use of specific antihypertensives, the utili-

zation of other frequently used medications, having as side-

effect hypotension (such as opioids), was not accounted

for, as well as procedures (such as ventriculostomies),

which may also affect the BP. Heart rate variability with

the nicardipine infusion was not assessed, because our

previous experience with the drug in other NICU patient

populations did not indicate any significant change in heart

rate. Lastly, although short-term functional outcomes were

not different than those previously reported in SAH [2],

long-term outcomes were not available.

Conclusion

Nicardipine is a safe and effective medication in controlling

acute hypertension or bringing the SBP at a pre-specified

goal after SAH, when other antihypertensive measures fail.

Although it often fails to maintain the SBP below these

goals, a refractory hypertensive profile of some patients, a

conservative titration protocol, a rather low SBP goal, and

the inability to completely control pain and agitation in these

patients may be in part responsible for this. We believe that a

larger study with more aggressive titration of nicardipine

will provide better results and bring more solid evidence on

its effectiveness on the SAH patient population.

Acknowledgments We would like to thank Lisa Pietrantoni and

Frances DeVos for their assistance in collecting the data. Dr. Varelas

and Dr. Abdelhak have received research grants from the Medicines

Company. Dr. Varelas also owns stocks of the Medicines Company

and has received Speaker honoraria by the same company. Funding:

Departmental supports.

References

1. Linn FH, Rinkel GJ, Algra A, van Gijn J. Incidence of sub-

arachnoid hemorrhage: role of region, year, and rate of computed

tomography: a meta-analysis. Stroke. 1996;27:625–9.

2. van Gijn J, Rinkel GJ. Subarachnoid haemorrhage: diagnosis,

causes and management. Brain. 2001;124:249–78.

3. Kassell NF, Torner JC. Aneurysmal rebleeding: a preliminary

report from the Cooperative Aneurysm Study. Neurosurgery.

1983;13:479–81.

4. Fujii Y, Takeuchi S, Sasaki O, Minakawa T, Koike T, Tanaka R.

Ultra-early rebleeding in spontaneous subarachnoid hemorrhage.

J Neurosurg. 1996;84:35–42.

5. Wallin JD, Fletcher E, Ram CV, et al. Intravenous nicardipine for

the treatment of severe hypertension. A double-blind, placebo-

controlled multicenter trial. Arch Intern Med. 1989;149:2662–9.

6. Adams HP Jr, del Zoppo G, Alberts MJ, et al. Guidelines for the

early management of adults with ischemic stroke: a guideline

from the American Heart Association/American Stroke Associ-

ation Stroke Council, Clinical Cardiology Council,

Cardiovascular Radiology and Intervention Council, and the

Atherosclerotic Peripheral Vascular Disease and Quality of Care

Outcomes in Research Interdisciplinary Working Groups: The

American Academy of Neurology affirms the value of this

guideline as an educational tool for neurologists. Circulation

2007;115:e478–534.

Neurocrit Care (2010) 13:190–198 197

7. Rosenbaum DM, Grotta JC, Yatsu FM, et al. Pilot study of nicar-

dipine for acute ischemic stroke. Angiology. 1990;41:1017–22.

8. Shah QA, Georgiadis A, Suri MF, Rodriguez G, Qureshi AI.

Preliminary experience with intra-arterial nicardipine in patients

with acute ischemic stroke. Neurocrit Care. 2007;7:53–7.

9. Qureshi AI, Harris-Lane P, Kirmani JF, et al. Treatment of acute

hypertension in patients with intracerebral hemorrhage using

American Heart Association guidelines*. Crit Care Med. 2006;34:

1975–80.

10. Powers WJ, Zazulia AR, Videen TO, et al. Autoregulation of

cerebral blood flow surrounding acute (6 to 22 hours) intracere-

bral hemorrhage. Neurology. 2001;57:18–24.

11. Narotam PK, Puri V, Roberts JM, Taylon C, Vora Y, Nathoo N.

Management of hypertensive emergencies in acute brain disease:

evaluation of the treatment effects of intravenous nicardipine on

cerebral oxygenation. J Neurosurg. 2008;109:1065–74.

12. Haley EC Jr, Kassell NF, Torner JC. A randomized trial of

nicardipine in subarachnoid hemorrhage: angiographic and

transcranial Doppler ultrasound results. A report of the Cooper-

ative Aneurysm Study. J Neurosurg. 1993;78:548–53.

13. Haley EC Jr, Kassell NF, Torner JC. A randomized controlled

trial of high-dose intravenous nicardipine in aneurysmal sub-

arachnoid hemorrhage. A report of the Cooperative Aneurysm

Study. J Neurosurg. 1993;78:537–47.

14. Haley EC Jr, Kassell NF, Torner JC, Truskowski LL, Germanson

TP. A randomized trial of two doses of nicardipine in aneurysmal

subarachnoid hemorrhage. A report of the Cooperative Aneurysm

Study. J Neurosurg. 1994;80:788–96.

15. Flamm ES, Adams HP Jr, Beck DW, et al. Dose-escalation study

of intravenous nicardipine in patients with aneurysmal sub-

arachnoid hemorrhage. J Neurosurg. 1988;68:393–400.

16. Lindegaard KF, Bakke SJ, Sorteberg W, Nakstad P, Nornes H. A

non-invasive Doppler ultrasound method for the evaluation of

patients with subarachnoid hemorrhage. Acta Radiol Suppl.

1986;369:96–8.

17. Farrell B, Godwin J, Richards S, Warlow C. The United Kingdom

transient ischaemic attack (UK-TIA) aspirin trial: final results.

J Neurol Neurosurg Psychiatry. 1991;54:1044–54.

18. Roitberg BZ, Hardman J, Urbaniak K, et al. Prospective ran-

domized comparison of safety and efficacy of nicardipine and

nitroprusside drip for control of hypertension in the neurosurgical

intensive care unit. Neurosurgery. 2008;63:115–20.

19. Liu-Deryke X, Janisse J, Coplin WM, Parker D Jr, Norris G,

Rhoney DH. A comparison of nicardipine and labetalol for acute

hypertension management following stroke. Neurocrit Care.

2008;9:167–76.

20. Broderick J, Connolly S, Feldmann E, et al. Guidelines for the

management of spontaneous intracerebral hemorrhage in adults:

2007 update: a guideline from the American Heart Association/

American Stroke Association Stroke Council, High Blood Pressure

Research Council, and the Quality of Care and Outcomes in

Research Interdisciplinary Working Group. Circulation. 2007;116:

e391–413.

21. Bederson JB, Connolly ES Jr, Batjer HH, et al. Guidelines for the

management of aneurysmal subarachnoid hemorrhage: a state-

ment for healthcare professionals from a special writing group of

the Stroke Council, American Heart Association. Stroke.

2009;40:994–1025.

22. Halpern NA, Goldberg M, Neely C, et al. Postoperative hyper-

tension: a multicenter, prospective, randomized comparison

between intravenous nicardipine and sodium nitroprusside. Crit

Care Med. 1992;20:1637–43.

23. Naidech AM, Janjua N, Kreiter KT, et al. Predictors and impact

of aneurysm rebleeding after subarachnoid hemorrhage. Arch

Neurol. 2005;62:410–6.

24. Ohkuma H, Tsurutani H, Suzuki S. Incidence and significance of

early aneurysmal rebleeding before neurosurgical or neurological

management. Stroke. 2001;32:1176–80.

25. Wijdicks EF, Vermeulen M, Murray GD, Hijdra A, van Gijn J.

The effects of treating hypertension following aneurysmal sub-

arachnoid hemorrhage. Clin Neurol Neurosurg. 1990;92:111–7.

26. Mayberg MR, Batjer HH, Dacey R, et al. Guidelines for the

management of aneurysmal subarachnoid hemorrhage. A state-

ment for healthcare professionals from a special writing group of

the Stroke Council, American Heart Association. Stroke.

1994;25:2315–28.

198 Neurocrit Care (2010) 13:190–198