CLINICAL STUDIES

Prophylactic Hyperdynamic Postoperative Fluid Therapy after

Aneurysmal Subarachnoid Hemorrhage: A Clinical,
Prospective, Randomized, Controlled Study

Arild Egge, M.D., Knut Waterloo, Ph.D.,

Hans Sjøholm, M.D., Ph.D., Tore Solberg, M.D.,
Tor Ingebrigtsen, M.D., Ph.D.,
Bertil Romner, M.D., Ph.D.
Department of Neurosurgery, University Hospital of Tromsø, Tromsø, Norway

OBJECTIVE: To investigate the role of prophylactic hyperdynamic postoperative fluid therapy in preventing delayed
ischemic neurological deficits attributable to cerebral vasospasm.
METHODS: We designed a prospected, randomized, controlled study and included 32 patients with subarachnoid
hemorrhage. Sixteen patients received hypervolemic hypertensive hemodilution fluid therapy; the other 16
patients received normovolemic fluid therapy. All patients were monitored for at least 12 days, with clinical
assessments, transcranial Doppler recordings, single-photon emission computed tomographic (SPECT) scanning,
and routine computed tomographic scanning. For fluid balance monitoring, a number of blood samples were
obtained on a daily basis and continuous central venous pressure and mean arterial blood pressure measurements
were performed for both groups. All patients received intravenous nimodipine infusions between Day 1 and Day
12. End points of this study were clinical outcomes, clinically evident and transcranial Doppler sonography-
evident vasospasm, SPECT findings, complications, and costs. Clinical examinations (using the Glasgow Outcome
Scale) performed 1 year after discharge, together with neuropsychological assessments and SPECT scanning, were
the basis for the evaluation of clinical outcomes.
RESULTS: No differences were observed between the two groups with respect to cerebral vasospasm (as observed
clinically or on transcranial Doppler recordings). When regional cerebral blood flow was evaluated by means of
SPECT analysis performed on Day 12 after subarachnoid hemorrhage, no differences were revealed. One-year
clinical follow-up assessments (with the Glasgow Outcome Scale), including SPECT findings and neuropsycho-
logical function results, did not demonstrate any significant group differences. Costs were higher and complica-
tions were more frequent for the hyperdynamic therapy group.
CONCLUSION: Neither early nor late outcome measures revealed any significant differences between the two
subarachnoid hemorrhage treatment models. (Neurosurgery 49:593–606, 2001)
Key words: Cerebral blood flow, Cerebral vasospasm, Hyperdynamic therapy, Induced hypertension, Neurological outcome, Subarachnoid
hemorrhage

O
utcomes for patients with aneurysmal subarachnoid
hemorrhage (SAH) have improved with advances in
neurosurgery and neurocritical care, including an
emphasis on early aneurysm clipping or coiling to prevent
rebleeding (19). However, cerebral ischemia related to vaso-
spasm remains an important cause of morbidity and death
(23), despite the beneficial effects of the calcium channel
blocker nimodipine (9). Intravascular volume contraction and
excessive natriuresis occur frequently after SAH and have
been implicated as risk factors for delayed ischemia related to
vasospasm (34, 59, 69).
During the past few decades, several reports from uncon-
trolled studies described the resolution of deficits resulting
from vasospasm after blood pressure elevation, volume ex-
pansion, and/or hemodilution (hypertensive hypervolemic
hemodilution therapy [“triple-H therapy”]) (3, 22, 25, 26, 29,
44), with improved outcomes with respect to vasospasm,
compared with historical control results. However, the effi-

Neurosurgery, Vol. 49, No. 3, September 2001 593

594 Egge et al.
cacy of triple-H therapy has not been demonstrated in con-
trolled trials, and studies of cerebral blood flow (CBF) after
the initiation of therapy have been equivocal (38). In addition,
studies have not been performed to determine which compo-
nent of this therapy is most important. Recently, Lennihan et
al. (28) found prophylactic hypervolemic therapy to be un-
likely to confer an additional benefit after SAH.
We performed this prospective, randomized, controlled
clinical study with 32 patients who were surgically treated
within 72 hours after SAH. The aim of the study was to
evaluate the role of hyperdynamic triple-H therapy in pre-
venting delayed ischemic neurological deficits after SAH.
PATIENTS AND METHODS
We performed a prospective, randomized, controlled study
between January 1997 and April 1999, with 32 patients who
were surgically treated within 72 hours after aneurysmal
SAH. All patients were in Hunt and Hess Grade I to III condition
(17). Patients in Hunt and Hess Grade IV or V condition were
excluded, because of late surgery (⬎72 h). Additional exclusion
criteria were age of more than 75 years, congestive heart failure,
pregnancy, and renal insufficiency. Aneurysmal SAH was doc-
umented by computed tomographic (CT) scanning, followed by
cerebral angiography, for all patients. The patients or relatives
were required to provide their written consent for participation,
and the regional research ethics committee approved the study.
Perioperative and intensive care unit management
All patients were treated according to a standard treatment
protocol. Mannitol (1 g/kg) was used during surgery for brain
relaxation, and standard microsurgical techniques were used
to clip the aneurysm and exclude it from the intracranial
circulation. Spinal drainage was not routinely used. Total
fluid input on the day of surgery ranged from 4.5 to 7 L for
most patients. Arterial cannulae and internal jugular venous
catheters were placed at the time of surgery. All patients
received nimodipine, administered intravenously, through-
out the study period. Phenytoin and dexamethasone were not
administered perioperatively. All patients were evaluated
hourly for signs of neurological deterioration. Transcranial
Doppler (TCD) sonography was performed on a daily basis.
Stratification and treatment randomization
Written consent was obtained on study Day 1. Prerandom-
ization stratification was performed according to the amount
of blood observed in the initial CT scans (Fisher grade) (10).
The Fisher grade was selected as a criterion because it is
expected to have major effects on vasospasm risks and clinical
outcomes. Prerandomization stratification on the basis of age,
sex, preexisting arterial hypertension, Hunt and Hess grade at
admission, or aneurysm location was not performed.
The included patients were distributed into two groups
(Groups A and B) and received fluid management in the
postoperative period according to two different protocols.
The first included patient in each Fisher grade was randomly
entered into one of the groups. Thereafter, consecutive pa-
tients were entered alternately into the two groups. The diag-
Neurosurgery, Vol. 49, No. 3, September 2001
nosis of clinical or TCD-detected vasospasm did not alter the
management. All included patients completed the study pro-
tocol, and there was no crossover between the two groups.
Fluid management
Group A patients (n ⫽ 16) received normovolemic fluid
therapy, including a baseline crystalloid infusion of 1000 ml of
5% dextrose and 1000 ml of 0.9% saline solution every day
until Day 12. Albumin solution or other colloids were not
administered. The aim of the normovolemic fluid manage-
ment (including free oral intake) was a neutral fluid balance.
Group B patients (n ⫽ 16) received triple-H fluid manage-
ment therapy. The fluid management strictly maintained the
central venous pressure (CVP) between 8 and 12 cm H2O,
venous hematocrit (Hct) between 30 and 35%, and mean
arterial blood pressure (MAP) more than 20 mm Hg greater
than the preoperative MAP value.
A baseline crystalloid infusion of 2000 ml of 5% dextrose
and 2000 ml of 0.9% saline solution and an infusion of 1000 to
1500 ml of colloids (500 ml of 4% albumin solution and/or
500–1000 ml of Rheomacrodex; Pharmalink AB, Spånga, Swe-
den) were administered on a daily basis until Day 12. Ap-
proximately two-thirds of the fluid infusion was administered
as crystalloids and one-third as colloids. Vasopressor admin-
istration (dopamine, 5–15 ␮g/kg/min) was titrated to yield
MAP values more than 20 mm Hg greater than the preoper-
ative MAP value.
All 32 patients received continuous intravenous nimodip-
ine infusions (0.2 mg/ml, 10 ml/h) for the entire study period
(Days 1 to 12), followed by oral administration (360 mg/d) for
10 to 14 days. Blood pressure monitoring was performed via
an arterial catheter.
We did not use pulmonary artery wedge pressures to guide
volume expansion because the nurses in the neurosurgical
ward were not certified to perform these measurements. Post-
operative angiography, papaverine infusion, and angioplasty
were not performed.
Early outcome measures
Physiological parameters
MAP and CVP were measured hourly, and median 24-hour
values were used for analysis. Hct, fluid intake, and fluid
balance were measured every 24 hours.
Scandinavian Neurological Stroke Scale assessments
The patients were monitored with daily clinical neurolog-
ical evaluations using a standardized scoring system, namely
the Scandinavian Neurological Stroke Scale (SNSS) (55).
TCD sonography
Daily TCD recordings were made for all patients (1). In-
creased flow velocities, indicating cerebral vasospasm, were
evaluated by using the index described by Lindegaard et al.
(31), based on the following formula: flow velocity in the
middle cerebral artery (MCA) divided by flow velocity in the
internal carotid artery (ICA). A value of more than 3 indicates
 Hyperdynamic Postoperative Fluid Therapy after SAH
vasospasm, and a value of more than 6 indicates severe va-
sospasm. A MCA/ICA ratio of more than 4 for 3 days or more
was defined as TCD-evident vasospasm in this study. TCD
recordings were conducted transtemporally using a 2-MHz
transducer, and examinations of the neck were performed
using a 4-MHz probe (DWL Multi Dop X; DWL, Sipplingen,
Germany).
CT scanning
Routine CT scanning was performed on Day 8 after SAH,
for evaluation of ventricular size, cerebral edema, and early
signs of ischemic lesions.
Single-photon emission computed
tomographic scanning
The early measures of treatment effect were assessed with
a mean of 2.7 (range, 1–5) single-photon emission computed
tomographic (SPECT) measurements on Days 4, 8, and 12
after SAH. The SPECT measurements performed on Day 12 after
SAH were used in the evaluation of early outcome measures.
Thirty-two patients were examined. CBF was measured with a
brain-dedicated imaging system (Siemens Neurofocal SPECT
camera; Siemens Medical Systems, Erlangen, Germany) with an
in-plane resolution of 12 mm. Each patient received an intrave-
nous injection of 750 MBq of 99mTc-hexamethylpropylamine
oxime (Ceretec; Amersham, Oslo, Norway) with eyes open dur-
ing and after the injection. Each patient was positioned in the
camera with the orbitomeatal line as the reference. The scanning
procedure lasted 20 minutes. Twelve transaxial images, 6.6 mm
thick (two pixels) and covering the whole brain, were recon-
structed using filtered back-projection and linear attenuation
correction. To perform quantitative estimations of the regional
CBF distribution, we first subtracted from all slices 30% of the
measured activity, which was considered general background
activity. The slices then had defined boundaries both laterally
and internally, compared with the ventricles, and covered essen-
tially cortical tissue. An automated template with 16 symmetri-
cal sectors was applied to all slices for computation of regional
activity. Side-to-side reductions in activity of 15% or more were
defined as abnormal CBF reductions. In an extensive SPECT
study in normal human subjects, Waldemar et al. (65) demon-
strated that CBF values in homologous bilateral brain regions
did not differ by more than 10%. In this study, we used the same
method but slightly different equipment. Therefore, we thought
it prudent to be somewhat more conservative, and we used a
15% side difference to distinguish between normal and abnor-
mal CBF values.
Evaluation of early outcome measures
Daily SNSS and TCD evaluations were performed without
blinding to the treatment assignments. CT and SPECT mea-
surements were performed and evaluated by clinicians
blinded to the treatment assignments of the study subjects.
Evaluation of late outcome measures
Late outcome measures of treatment effects included clin-
ical examinations performed 1 year after SAH, using the
Neurosurgery, Vol. 49, No. 3, September 2001
595
Glasgow Outcome Scale (GOS) (20). This examination was
performed by one of the authors (AE), who was thus not
blinded to the treatment assignments of the study subjects. On
the same day, SPECT measurements and neuropsychological
assessments were performed. Investigators blinded to the
treatment assignments of the study subjects performed these
analyses.
Neurobehavioral assessments
For neuropsychological testing, only standardized, com-
monly used tests were applied. They were chosen to assess
cognitive dysfunctions usually observed after diffuse brain
damage as well as focal damage, because SAH involves the
likelihood of diffuse as well as more localized disruption of
brain cortices. One year after SAH, all patients underwent a
battery of seven main conventional tests and four computer-
ized tests. The areas of neuropsychological functioning exam-
ined and the specific tests administrated were as follows: 1)
attention: digit span test from the Wechsler Adult Intelligence
Scale (66) and Seashore Rhythm Test (52); 2) psychomotor
speed: trail-making test, Parts A and B (52), and Stroop Color-
Word Test (reading speed) (60, 61), modified version (12, 33);
3) memory: immediate recall and 30-minute delayed recall of
two subtests from the Wechsler Memory Test-Revised (67),
namely the verbal paired associates and visual paired associ-
ates subtests; 4) language (word fluency): Controlled Oral
Word Association test (30, 60); 5) cognitive flexibility: Stroop
Color-Word Test (interference) (60, 61) and a computerized
version of the Wisconsin Card Sorting Test (14); 6) speed of
information-processing: California Computerized Assessment
Package, abbreviated version, containing a measure of simple
reaction time and three measures of complex choice reaction
time (41). Furthermore, to evaluate intellectual functions, the
similarities (verbal function), block design (visuospatial con-
struction), and picture arrangement (nonverbal functions)
subtests from the Wechsler Adult Intelligence Scale were used
(66). Mood was measured with the Beck Depression Inventory
(BDI), which was included because of the confounding effects
depressed mood can have on cognitive performance (4). The
BDI is a widely used depression questionnaire with good
reliability and validity (68).
A trained test technician, blinded regarding to which group
the patients belonged, conducted all examinations. All pa-
tients were tested individually. The technician scored all tests,
after which they were sent to another technician, who re-
viewed the findings for accuracy. An experienced neuropsy-
chologist (KW), also blinded regarding to which group the
patients belonged, coded the test scores and performed data
analyses.
The 2.5- to 3-hour neuropsychological test battery was ad-
ministrated in two sessions. Two patients were lost to
follow-up monitoring, and the follow-up examinations in-
cluded 30 patients. Factors known to influence neuropsycho-
logical performance, including drug or alcohol abuse, learn-
ing disabilities, and previous psychiatric or neurological
trauma or disease, were analyzed by review of patient
596 Egge et al.
records; there was no history of these factors among the
patients with SAH.
Complications
All complications observed during the study period were
prospectively recorded on a daily basis.
Costs
Additional costs for fluid management with triple-H ther-
apy were calculated. Additional costs for patient treatment
with triple-H therapy (e.g., prolonged hospital stays, evacua-
tion of postoperative hematomas, and additional nursing
work) were not taken into account.
Statistical analyses
Results are presented as means and standard deviations.
Data analyses included Student’s unpaired t test (two-tailed)
or analysis of variance to test differences between group
means and differences between two or more groups of quan-
titative data and the ␹2 test to test differences among groups
of qualitative data. Because of multiple comparisons, a
Bonferroni-corrected significance threshold of P ⬍ 0.01 was
considered appropriate, to reduce the risk of Type I errors.
Simple and multiple regressions were used to analyze
quantitative factors associated with neuropsychological pa-
rameters, with the test score as the continuous dependent
variable and demographic and SAH-associated factors as the
independent variables. Differences between groups and times
in repeated physiological measures (MAP, CVP, and Hct)
were assessed using repeated-measures analysis of variance.
Differences were considered significant at P ⱕ 0.05. Data were
analyzed using SPSS for Windows (Release 10.0; SPSS Inter-
national BV, Gorinchem, The Netherlands) and Statview
Graphics (Version 5.0; SAS Institute, Inc., Cary, NC).
RESULTS
Early outcome measures
Physiological variables
Table 1 presents baseline characteristics such as age, sex,
Hunt and Hess grade at admission, Fisher grade at admission,
TABLE 1. Baseline Characteristicsa
b c
Age (yr) Hunt and Hess Grade
Sex (M/F)
1 2 3 4
Group A 48.6 (28–65) 10/6 11 2 3 0 0
Group B 47.4 (28–73) 9/7 11 4 1 0 0
a
Group A, 16 patients who received normovolemic therapy for 12 days after aneurysmal subarachnoid hemorrhage (SAH); Group B, 16
patients who received hyperdynamic therapy for 12 days after aneurysmal SAH. AComA, anterior communicating artery; PComA, posterior
communicating artery; MCA, middle cerebral artery.
b
Values represent group means, and ranges.
c
Hunt and Hess grade at admission (17).
d
Amount of blood (Fisher grade) evaluated on the initial computed tomographic scans (10).
Neurosurgery, Vol. 49, No. 3, September 2001
and aneurysm location. Fisher grades were (according to the
randomization strategy) equal in the two groups. There was
no significant difference in Hunt and Hess grades at admis-
sion between the groups. Aneurysm locations were equal in
the two groups.
Physiological parameters (MAP, CVP, and Hct) are pre-
sented in Table 2 and Figures 1 to 3. Median 24-hour MAP and
CVP values were used for all patients. MAP was significantly
higher for Group B patients on Days 2 to 12 (P ⫽ 0.001).
Induced hypertensive therapy (dopamine, 5–15 ␮g/kg/min)
was used for Group B patients, to obtain MAP values more
than 20 mm Hg greater than preoperative values. No patient
in Group A underwent induced hypertensive therapy.
Mean CVP values for Group B patients were 9.7 ⫾ 2.1 cm
H2O, compared with 7.2 ⫾ 2.8 cm H2O for Group A patients.
The difference with respect to Group A patients was statisti-
cally significant between Day 5 and Day 12 (P ⫽ 0.005). The
CVP values for Group A patients were within the normal
range (5–10 cm H2O) throughout the study period (Fig. 2).
Hct levels were significantly lower for Group B patients
between Day 5 and Day 12 (P ⫽ 0.03) (Table 2). As demon-
strated in Figure 3, no significant initial difference was ob-
served, and the Hct levels were decreased for both groups on
Days 1 to 3. Fluid administration differed between the groups,
with significantly greater fluid intake for Group B (P ⬍ 0.0001)
(Fig. 4).
Clinical vasospasm evaluated by daily
SNSS assessments
The severity of clinical symptoms did not differ signifi-
cantly (Fisher exact test) between the two groups. Five pa-
tients (31.5%) in Group A and four patients (25%) in Group B
demonstrated clinical symptoms of vasospasm (Table 3). The
clinical symptoms were reduced levels of consciousness (two
patients in Group A and two patients in Group B) and hemi-
paresis (three patients in Group A and two patients in Group
B). Appearance of headache during the study period was not
considered a symptom of vasospasm.
TCD sonography
Ten patients in each group demonstrated MCA/ICA ratios
of more than 4 for more than 3 days, indicating TCD-evident
No. of Patients
Fisher Graded Aneurysm Location
5 1 2 3 4 AComA PComA MCA Other
0 8 7 1 4 4 7 1
0 8 7 1 4 5 7 0
 Hyperdynamic Postoperative Fluid Therapy after SAH 597

TABLE 2. Physiological Parametersa

Group A Group B
Admittance Days 2–12 Admittance Days 2–12

MAP (mm Hg) 86.0 ⫾ 10.3 87.6 ⫾ 7.8 81.8 ⫾ 6.9 103.7 ⫾ 10.1b
Hct (%) 37.4 ⫾ 3.9 33.4 ⫾ 3.7 35.1 ⫾ 5.1 30.3 ⫾ 3.6c
CVP (cm H2O) 7.2 ⫾ 2.8 9.7 ⫾ 2.1d
a
Mean ⴞ standard deviation values for mean arterial blood pressure (MAP), hematocrit (Hct) levels, and central venous pressure (CVP) were
determined at admission and on Days 2 to 12. The Hct levels were significantly lower for Group B patients between Day 5 and Day 12 (P ⴝ 0.03).
Group A, 16 patients who received normovolemic therapy for 12 days after aneurysmal subarachnoid hemorrhage (SAH); Group B, 16 patients
who received hyperdynamic therapy for 12 days after aneurysmal SAH.
b
P ⴝ 0.001.
c
P ⴝ 0.03.
d
P ⴝ 0.01.

FIGURE 3. Plot demonstrating mean Hct levels for Group A

FIGURE 1. Plot demonstrating MAP values for Group A (□) (□) and Group B (䡵) patients during the 12-day study
and Group B (}) patients during the 12-day study period. period. Hct values were significantly lower for Group B
MAP values were significantly higher for Group B patients on patients between Day 5 and Day 12 (P ⴝ 0.03).
Days 2 to 12 (P ⴝ 0.001).

FIGURE 4. Plot demonstrating mean daily fluid intake for

Group A (□) and Group B (䡵) patients during the 12-day
FIGURE 2. Plot demonstrating mean CVP values for Group
study period. The fluid administration differed between the
A (□) and Group B (䡵) patients during the 12-day study
groups, with significantly greater fluid intake for Group B
period. CVP values were significantly higher for Group B
(P < 0.0001).
patients (P ⴝ 0.005).

vasospasm (Table 3). Six subjects in each group exhibited no TCD-evident vasospasm for more than 3 days without clinical
TCD-evident vasospasm (MCA/ICA ratios of ⬍4). Five pa- deterioration. Flow velocities reached the highest levels dur-
tients in Group A and six patients in Group B exhibited ing the second week after SAH.

Neurosurgery, Vol. 49, No. 3, September 2001

598 Egge et al.
TABLE 3. Clinical Vasospasm (Evaluated by Daily
Scandinavian Neurological Stroke Scale Scores) and
Transcranial Doppler-evident Vasospasma
No. of Patients
Group A Group B
Both clinical and TCD-evident 5 4
vasospasm
TCD-evident but not clinical 5 6
vasospasm
No TCD-evident vasospasm, no 6 6
clinical vasospasm
Total 16 16
a
TCD, transcranial Doppler sonography (1), as evaluated by using
the Lindegaard index (31). Group A, 16 patients who received normo-
volemic therapy for 12 days after aneurysmal subarachnoid hemor-
rhage (SAH); Group B, 16 patients who received hyperdynamic ther-
apy for 12 days after aneurysmal SAH.
SPECT measurements
No mean group differences were demonstrated when
SPECT measurements obtained on Day 12 after SAH were
compared (Table 4). In both groups, signs of reduced regional
CBF were present. There were large variations in regional CBF
reductions in both groups. Early SPECT measurements re-
vealed clear signs of hyperemia for four patients in Group A
and eight patients in Group B.
Early clinical outcomes
All patients were evaluated 14 days after SAH by means of
the GOS (Table 5). All patients in Group A exhibited early
favorable outcomes (GOS scores of 4 or 5), whereas three
patients in Group B were considered severely disabled (GOS
scores of 3).
Late outcome measures
There were no statistically significant differences in age or
level of education between the two patient groups. Ages in
Group A ranged from 28 to 65 years, with a mean of 48.6
years. Ages in Group B ranged from 28 to 73 years, with a
mean of 47.4 years. The mean educational level was 9.5 ⫾ 2.6
years for Group A and 10.3 ⫾ 3.1 years for Group B.
One-year clinical follow-up results
Table 5 presents outcomes, according to GOS scores, 1 year
after SAH. There were no significant differences between the
groups. One patient died in each group. The cause of death for
the patient in Group B was pulmonary embolism. The patient
in Group A died shortly after discharge from our clinic,
probably because of cardiac arrhythmia; the autopsy report
was inconclusive. Three patients were severely disabled. The
unfavorable outcome was attributable to vasospasm for two
patients, whereas one patient (in Hunt and Hess Grade III at
admission) experienced worsening after surgery, most likely
because of temporary clipping during surgery.
Neurosurgery, Vol. 49, No. 3, September 2001
SPECT measurements
No mean group differences were detected when SPECT
measurements obtained 1 year after SAH were compared
(Table 4). Compared with early SPECT measurements, the
volumes of the regional CBF reductions were significantly
increased in both groups (Group A, P ⫽ 0.006; Group B, P ⫽
0.02). Only a few patients in the two groups demonstrated
normal regional CBF distribution.
Neurobehavioral outcomes
There were no statistically significant differences in perfor-
mance between the two patient groups exceeding the 0.01
level, as summarized in Table 6. Furthermore, there were no
group differences between mean scores on the BDI. Seventy-
one percent of BDI scores for patients in Group A were within
the not-depressed range (scores of 0–9), whereas 29% of pa-
tients exhibited scores in the mild-to-moderate depression
range (scores of 10–19). For patients in Group B, 64% of the
scores were within the normal range (scores of 0–9), whereas
36% of the scores were between 10 and 19. There was no
significant association between BDI scores and the scores on
any individual neuropsychological test.
In multiple-regression equations with neuropsychological
test scores as the dependent variables and the neurological
(Hunt and Hess) grade at admission, severity of the initial
bleeding (Fisher grade), vasospasm, age, and education as
independent variables, the effects on the Stroop test (color-
word interference) results were observed to be attributable to
age (b ⫽ 3.0, P ⫽ 0.02). The performance on the complex
reaction time test (California Computerized Assessment Pack-
age) was significantly influenced by the severity of the initial
bleeding (Fisher score) (b ⫽ 141.3, P ⫽ 0.004). No other de-
mographic factor or disease-associated factor had significant
effects on the performance on any test.
Complications
There were significant differences in the frequencies of
complications between the treatment groups (P ⬍ 0.001). Most
of the complications (82%) were observed for Group B pa-
tients during the study period (Table 7). Two patients in
Group B developed acute extradural hematomas, and one
required surgical intervention. Three patients, all in Group B,
experienced coagulation failure, leading to a tendency for
bleeding complications at cannula sites and wounds and pro-
longed menstruation.
Congestive heart failure combined with arrhythmia and
pulmonary edema occurred in two Group B patients. Local
infections (cannulae or wounds) occurred in a total of four
patients, two in each group.
Costs
The additional costs for hypervolemic therapy (including
induced hypertension), compared with normovolemic ther-
apy, were a mean of $250/d.
 Hyperdynamic Postoperative Fluid Therapy after SAH 599

TABLE 4. Volume of Cerebral Blood Flow Reduction, as Measured with 99mTc-Hexamethylpropylamine Oxime Single-photon
Emission Computed Tomography, 12 Days and 12 Months after Subarachnoid Hemorrhagea
Volume of CBF Decrease No. of Patients with Hyperemia Volume of CBF Decrease 12
Group
on Day 12 after SAH on Day 12 after SAHb mo after SAHc

A 3.5 (range, 0–16) 4 10.1 (range, 0–38)

B 3.1 (range, 0–29) 8 11.9 (range, 0–49)
a
CBF, cerebral blood flow. Group A, 16 patients who received normovolemic therapy for 12 days after aneurysmal subarachnoid hemorrhage
(SAH); Group B, 16 patients who received hyperdynamic therapy for 12 days after aneurysmal SAH. For number of cortical sectors with CBF
decrease, see Patients and Methods.
b
Number of patients demonstrating hyperemia in single-photon emission computed tomographic (SPECT) measurements. No significant group
differences were demonstrated.
c
No significant group differences were detected. Compared with early SPECT measurements, the volume of the regional CBF reductions was
significantly increased in both groups (Group A, P ⴝ 0.006; Group B, P ⴝ 0.02).

TABLE 5. Clinical Outcomesa Stratification and treatment randomization

No. of Patients
14 d after SAH 12 mo after SAH
Group A Group B Group A Group B
Good recovery 10 11 12 9 truly reliable prognostic factors in aneurysmal SAH (10, 54).
Moderately disabled 6 2 2 4
Severely disabled 0 3 1 2
Persistent vegetative state 0 0 0 0
Dead 0 0 1 1
a
Clinical outcomes were evaluated 14 days and 1 year after sub-
arachnoid hemorrhage (SAH), using the Glasgow Outcome Scale (20).
Group A, 16 patients who received normovolemic therapy for 12 days
after aneurysmal SAH; Group B, 16 patients who received hyperdy-
namic therapy for 12 days after aneurysmal SAH.
DISCUSSION
Hypervolemic therapy is now routinely used after surgery
at most medical centers, with the assumption that this inter-
vention may increase CBF, prevent delayed ischemia, and
improve clinical outcomes (40, 57, 58). However, both in-
creases (43, 47) and decreases (70) in CBF have been reported
after volume expansion among patients with SAH. An uncon-
trolled series suggested that therapy might be more effective
if initiated prophylactically before the onset of symptoms,
preferably after aneurysm clipping (57). In most centers,
triple-H treatment is usually continued beyond the period of
risk for vasospasm or until the reduction of vasospasm, as
assessed by clinical and TCD parameters.
This randomized controlled study evaluated triple-H ther-
apy as prophylactic treatment for vasospasm after SAH. Both
Group A and B patients were treated according to the protocol
for 12 days, with daily neurological examinations and TCD
measurements to detect eventual signs of vasospasm. One
major drawback of this study was the small number of pa-
tients included. Furthermore, this study does not answer the
question of whether triple-H treatment is effective for symp-
tomatic clinical vasospasm, for which it is mostly
recommended.
A stratified randomization was performed on the basis of
the amount of blood (Fisher grade) evident on the initial CT
scans. The strong relationship between the amount of extrav-
asated blood and the final outcome seems to be one of the few
The clinical syndrome of vasospasm was elaborated in detail
by Fisher et al. (11). Kapp et al. (21) established that the
development of vasospasm and its clinical significance could
be clearly predicted by early CT scans. In brief, severe cerebral
vasospasm with clinical manifestations occurs mostly among
patients with thick blood clots in the basal cisterns, and the
vasospasm is more severe where the blood clots are the
thickest.
End points
Physiological variables
The goal of increasing MAP by more than 20 mm Hg,
compared with preoperative levels, was achieved, as demon-
strated in Figure 2. This significant group difference was ob-
served between Day 2 and Day 12, demonstrating a clear
difference in postoperative hyperdynamic treatment
strategies.
In this study, the fluid management (hypervolemic treat-
ment) was strictly followed to achieve CVP values between 8
and 12 cm H2O for triple-H patients (Group B). We were able
to achieve these values for all Group B patients during the
study period. The difference with respect to Group A patients
was statistically significant between Day 5 and Day 12. The
CVP values for Group A patients decreased after study Day 4
but were still within the normal range.
Hct measurements demonstrated a statistically difference
between the two groups. After the initial decrease for both
groups between Day 1 and Day 3, the Hct levels remained
stable for Group A patients, whereas a daily decrease was
observed for Group B patients. Ekelund et al. (7) retrospec-
tively evaluated hemodilution for 36 patients who were sur-
gically treated for aneurysmal SAH, and they observed no
difference between active hemodilution therapy and normal
postoperative treatment.

Neurosurgery, Vol. 49, No. 3, September 2001

600 Egge et al.

TABLE 6. Neuropsychological Function 12 Months after Subarachnoid Hemorrhagea

Tests Group A (n ⫽ 15) Group B (n ⫽ 15) P Values

Intellectual function
Verbal function
WAIS, similarities1,b 50.3 ⫾ 8.9 57.4 ⫾ 5.2 0.02
Nonverbal function
WAIS, picture arrangement1,b 44.8 ⫾ 10.1 43.8 ⫾ 9.5 0.8
WAIS, block design1,b 49.7 ⫾ 12.4 47.2 ⫾ 13.3 0.6
Cognitive flexibility
WCST, persev. responses1,b 37.4 ⫾ 9.3 38.1 ⫾ 10.7 0.9
WCST, % concept level1,b 37.3 ⫾ 8.4 37.4 ⫾ 9.9 1.0
Stroop test (s) 101.4 ⫾ 60.0 103.0 ⫾ 74.0 0.9
Language
COWA, total words2,b 20.1 ⫾ 10.2 24.1 ⫾ 11.7 0.3
Psychomotor speed
Trail-making test Part A1,b 38.5 ⫾ 8.9 37.3 ⫾ 8.0 0.7
Trail-making test Part B1,b 37.6 ⫾ 8.1 37.1 ⫾ 11.5 0.9
Memory
WMS-R, verbal paired ass., immediate 2,b 12.5 ⫾ 5.5 15.4 ⫾ 4.6 0.1
WMS-R, verbal paired ass., delayed 2,b 4.7 ⫾ 2.1 5.6 ⫾ 2.1 0.3
WMS-R, visual reprod, immediate 2,b 8.7 ⫾ 3.7 9.5 ⫾ 5.1 0.7
WMS-R, visual reprod, delayed 2,b 3.2 ⫾ 2.1 2.9 ⫾ 2.3 0.8
Attention
Digit span test1,b 40.8 ⫾ 7.8 35.9 ⫾ 6.7 0.08
Seashore Rhythm Test1,b 41.9 ⫾ 12.0 35.5 ⫾ 8.9 0.1
Speed of information processing
CalCAP, complex RT, digits (ms) 546.9 ⫾ 139.7 573.3 ⫾ 168.4 0.6
CalCAP, complex RT, Seq. 1 (ms) 658.8 ⫾ 122.7 704.0 ⫾ 147.9 0.4
CalCAP, complex RT, Seq. 2 (ms) 736.7 ⫾ 132.8 773.3 ⫾ 169.6 0.5
Mood self-evaluation
BDI score2 5.5 ⫾ 3.7 7.1 ⫾ 4.4 0.3
a
ass, associates; reprod, reproduction; WMS-R, Wechsler Memory Scale-Revised; CalCAP, California Computerized Assessment Package; RT,
reaction time; Seq, sequential reaction time; Stroop test, Stroop Color-Word Test; WCST, Wisconsin Card Sorting Test; persev, perseverative;
WAIS, Wechsler Adult Intelligence Scale; BDI, Beck Depression Inventory; COWA, Controlled Oral Word Association test. Group A, 16 patients
who received normovolemic therapy for 12 days after aneurysmal subarachnoid hemorrhage (SAH); Group B, 16 patients who received
hyperdynamic therapy for 12 days after aneurysmal SAH. The results are given as mean ⴞ standard deviation. The test results are presented as
standardized t scores (1) or raw scores (2).
b
Higher scores indicate better performance.

In 1998, Mayer et al. (39) reported a study of 43 patients
with aneurysmal SAH who were treated with two different
fluid protocols. For one group of patients, albumin was ad-
ministered in sufficient amounts to maintain the CVP at ap-
proximately 8 mm Hg. For the second group, the CVP was
maintained at approximately 5 mm Hg. Supplemental 5%
albumin administered to maintain the CVP above 8 mm Hg
prevented sodium and fluid losses but did not have an effect
on blood volume. Blood volume was decreased by 10% in
both groups.
Lennihan et al. (28) evaluated the effect of hypervolemic
therapy on CBF after SAH for 82 patients. Those authors used
a stratified treatment randomization including the number of
days since SAH and the postoperative Hunt and Hess grade,
whereas other factors that have been predictive of delayed
ischemia, such as the amount of SAH on admission CT scans
or TCD data, were not taken into account. CBF values for
hypervolemic and normovolemic subjects were compared.
Hypervolemic patients received significantly more fluid and
exhibited higher pulmonary artery diastolic pressure and
CVP, compared with normovolemic patients. There was no
difference in mean global CBF values during the treatment
period between the groups. Symptomatic vasospasm oc-
curred in 20% of the patients in each group and was associ-
ated with reduced minimal regional CBF values (P ⫽ 0.04).
Patients receiving triple-H treatment are usually monitored
in an intensive care unit with a Swan-Ganz catheter, arterial
line, and frequent serum electrolyte determinations. In many

Neurosurgery, Vol. 49, No. 3, September 2001

 Hyperdynamic Postoperative Fluid Therapy after SAH
TABLE 7. Complications during the Entire Study Period for
Groups A and Ba
No. of Patients
Group A Group B
Extradural hematomas 0 2
Hemorrhagic diatesis 0 3
Congestive heart failure/arrhythmia 0 2
Infectious complications 2 2
a
There were significant differences in the frequency of complica-
tions between the two treatment groups (P < 0.001). Group A, 16
patients who received normovolemic therapy for 12 days after aneu-
rysmal subarachnoid hemorrhage (SAH); Group B, 16 patients who
received hyperdynamic therapy for 12 days after aneurysmal SAH.
protocols, measurements of left ventricular end diastolic pres-
sure and cardiac output are used to optimize hemodynamics
according to the Starling curve (29). Unfortunately, we were
not able to obtain pulmonary artery wedge pressure measure-
ments to guide volume expansion, because our neurointen-
sive care nurses were not certified to perform these
measurements.
Muizelaar and Becker (44) retrospectively evaluated in-
duced arterial hypertension among patients who postopera-
tively developed clinical signs of cerebral ischemia attribut-
able to vasospasm. Using intravenous phenylephrine
treatment, they documented an effect of such treatment on
CBF measurements, combined with an immediate and obvi-
ous positive clinical effect, for all patients. Darby et al. (6)
observed that dopamine-induced hypertension led to in-
creased CBF in ischemic noninfarcted territories, without pro-
ducing an increase in mean global CBF. Mizuno et al. (42)
administered prophylactic hyperdynamic therapy (including
induced hypertension) to most patients and observed stable
CBF values within 3 weeks after SAH. We used hypervolemic
therapy (including induced hypertension) as a prophylactic
intervention for vasospasm but were not able to confirm those
findings. However, our study included only 16 patients re-
ceiving triple-H therapy, and the dopamine dose adminis-
tered was low (5–15 ␮g/kg/min). With the use of induced
hypertension, we aimed to increase MAP by more than 20
mm Hg during the entire study period, compared with pre-
operative MAP values. We also achieved a significant differ-
ence in MAP values between the two study groups. However,
additional increases in MAP could be more effective prophy-
lactic therapy.
Clinical vasospasm evaluated by daily
SNSS assessments
There was no difference in the frequency or severity of
clinical symptoms of vasospasm between the groups. Five
patients in Group A and four patients in Group B demon-
strated clinical symptoms of vasospasm.
Neurosurgery, Vol. 49, No. 3, September 2001
601
TCD assessments
TCD-evident vasospasm is one important factor that has
been predictive of delayed ischemia after aneurysmal SAH.
Flow velocities of more than 120 cm/s in the MCA and daily
increases in MCA flow velocity of more than 50 cm/s have
been established as indicators of cerebral vasospasm (8, 13).
However, Vora et al. (64) recently concluded that, for individ-
ual patients, only low (⬍120 cm/s) or very high (⬎200 cm/s)
MCA flow velocities reliably predicted the absence or pres-
ence of clinically significant vasospasm. With the use of the
MCA/ICA ratio (Lindegaard index), a more accurate evalua-
tion of increased flow velocities can be performed (31). Hy-
peremia can be differentiated from vasospasm; a ratio of more
than 3 indicates vasospasm, whereas a ratio of more than 6
indicates severe vasospasm. We observed no differences be-
tween the two groups with respect to TCD-evident vaso-
spasm during the study period. In both groups, a ratio of
more than 4 was demonstrated for 10 patients. A ratio of less
than 3 (no TCD-evident vasospasm) was observed for six
patients in each group. Five patients in Group A and six pa-
tients in Group B exhibited TCD-evident vasospasm but no
clinical signs of vasospasm.
Early clinical outcomes
All patients in Group A exhibited early favorable outcomes
(GOS scores of 4 or 5), whereas three patients in Group B were
considered severely disabled (GOS scores of 3).
SPECT scanning
Experimental studies indicated that volume expansion may
improve CBF in ischemic regions independently of perfusion
pressure, because of beneficial effects on cardiac output and
blood rheology (24, 36, 62), and uncontrolled case series re-
ported that hypervolemic therapy could reverse ischemic def-
icits among patients with symptoms (3, 22, 26, 51). SPECT
scanning provides an estimate of regional CBF or global CBF,
which (with a few exceptions) is closely associated with brain
metabolism (35). A major exception involves patients with
focal ischemia. SPECT scanning with hexamethylpropylamine
oxime is not a measure of absolute CBF but is well suited for
the detection of regional CBF abnormalities (35). No mean
group differences in SPECT measurements were demon-
strated either 12 days or 1 year after SAH. Interestingly, when
early SPECT measurements were compared with late mea-
surements, the volume of the regional CBF reductions was
significantly increased in both groups. Both Group A and B
patients with normal CBF distribution on Day 12 after SAH
exhibited large regional CBF reductions 1 year later. Regional
hyperemia was present in both Group A and B patients up to
12 days after SAH, probably as a sign of “luxury perfusion”
(27). This is a phenomenon linked to unstable CBF, and it
often appears before later infarctions (46).
Neurobehavioral outcomes
Neurobehavioral assessment provides an objective and
sensitive method for delineating changes in cerebral functions
and has been demonstrated to be a useful diagnostic proce-
602 Egge et al.
dure for assessment of organic brain dysfunction (30, 52).
Although many survivors of SAH experience excellent neu-
rological recoveries, several studies of cognitive outcomes
noted that a large proportion do not fully regain their pre-
morbid status (5, 18, 32, 37). Memory dysfunction after SAH
has attracted particular attention, both because of the impor-
tance of memory as a factor in the patient’s daily life and
because of the many studies reporting long-term memory
dysfunctions after SAH (5, 18, 32). Although memory prob-
lems seem to be the most frequent cognitive deficits resulting
from SAH, other areas of cognition also suffer; dysfunctions
in attention, psychomotor and information-processing speed,
and cognitive flexibility have been reported (5, 18, 32).
No global cognitive disturbances were observed and no sig-
nificant group differences in cognitive function were observed
for patients who received hypervolemic fluid therapy according
to the triple-H model, compared with patients who received
normovolemic fluid therapy. The neurological grade at admis-
sion (Hunt and Hess grade) could not predict the extent of
cognitive dysfunction 1 year after SAH, whereas the severity of
the initial bleeding (Fisher grade) had a significant effect on the
speed of information-processing, which is in agreement with
other studies (18). The age of the patients at the time of SAH
proved to be a significant predictor of cognitive flexibility dys-
function. In 12-month assessments after SAH, older patients
were significantly more disturbed in this area of cognitive func-
tioning than were younger patients. Similar results were re-
ported by Hütter and Gilsbach (18) and Bornstein et al. (5).
No other demographic factor or disease-associated factor had
significant effects on performance on any test.
Late clinical outcome measures
It is a commonly held opinion that some factors affect
outcomes more than others for patients with aneurysmal
SAH. Greater age, poor clinical condition at admission, his-
tory of arterial hypertension, low CBF, and greater amounts of
blood on the initial CT scans have all been considered nega-
tive prognostic factors (2, 16, 43, 45, 49, 53, 54). The Hunt and
Hess grade at admission did not differ between the groups in
this study. Clinical outcomes, as assessed by means of GOS
scores measured 12 months after SAH, indicated 14 patients
with favorable outcomes in Group A and 13 patients with
favorable outcomes in Group B. The rates of unfavorable
outcomes were similar for the two groups. One patient in each
group died before the late follow-up examination.
Complications
Eighty-two percent of the complications during the study
period occurred among Group B patients. This difference in
frequency was statistically significant (P ⬍ 0.001). The most
serious complications were extradural hematomas (observed
for two patients), which required surgical evacuation in one
case. These had no effects on outcomes. Further complications
were treated conservatively; however, complications pro-
longed the hospital stay for four patients.
Shimoda et al. (56) retrospectively concluded that hypervo-
lemic therapy might be very harmful in the early phase after
Neurosurgery, Vol. 49, No. 3, September 2001
SAH. Among patients who received hypervolemic therapy, 68
patients (72%) experienced no intracranial complications,
whereas 18 (19%) exhibited aggravation of brain edema and 8
(9%) developed hemorrhagic infarctions. Trumble et al. (63)
evaluated the use of colloidal agents to achieve hypervolemia
for the prevention and treatment of vasospasm after SAH.
Among a series of 85 patients with recent aneurysmal SAH, 26
developed clinical symptoms of vasospasm. Fourteen of the
26 patients were treated with hetastarch for volume expan-
sion, whereas the other 12 received plasma protein fraction.
Clinically significant, bleeding-associated pathological condi-
tions were noted for six patients who received hetastarch as a
continuous intravenous infusion. Hetastarch increased partial
thromboplastin times for all patients who received this agent,
whereas no effect was noted for the 12 patients who received
plasma protein fraction infusions.
Costs
A comparison of costs for Group A and B patients demon-
strated a mean additional cost of $250/24 h for the fluid
management. When used, triple-H therapy is administered
for a period of 7 to 14 days at most centers. Therefore, 10 days
of triple-H therapy would yield an additional fluid cost of at
least $2500.
Practical aspects
In a recent review of hyperdynamic therapy, Pritz (50)
presented practical aspects of the treatment of cerebral vaso-
spasm. Successful medical treatment of neurological symp-
toms attributable to vasospasm requires vigilance, accurate
diagnoses, and prompt intervention. The results of TCD as-
sessments performed routinely at regular intervals after SAH,
coupled with neurological decline in the absence of electrolyte
and/or cardiovascular or pulmonary abnormalities, should
suggest that these symptoms result from vasospasm. CT scans
can exclude cerebral edema, hydrocephalus, stroke attribut-
able to vessel occlusion, and hematomas as the causes of these
deficits. Cerebral angiography can document the vessels af-
fected by vasospasm, as well as satisfactory aneurysm clip-
ping and vessel patency.
There has been an extraordinary amount of laboratory work
studying the pathogenesis and possible pharmacological
treatment of vasospasm, but no treatment has been proven to
be useful in clinical practice (15). Recently, Polin et al. (48)
raised significant doubts regarding the long-term effective-
ness of endovascular therapy for vasospasm.
However, only a proportion of patients with vasospasm
respond to triple-H therapy, with vasospasm-related stroke
and death rates approaching 15% in the series with the best
outcomes (3, 44). Initiation of triple-H therapy is associated
with significant risks, including the possibilities of cardiac
failure, electrolyte abnormalities, cerebral edema, bleeding
abnormalities, and rupture of unsecured aneurysms (29).
These complications were confirmed in our study. Furthermore,
we found triple-H therapy to be associated with higher costs. In
summary, we observed that prophylactic triple-H therapy after
aneurysm clipping did not result in increased regional CBF (as
 Hyperdynamic Postoperative Fluid Therapy after SAH
assessed by means of SPECT scanning), less clinically evident or
TCD-evident vasospasm, or differences in neurocognitive im-
pairment, compared with normovolemic therapy.
CONCLUSION
This study does not support the idea that hyperdynamic
postoperative fluid therapy (triple-H therapy) prevents de-
layed ischemic neurological deficits, compared with normo-
volemic fluid therapy, among patients with SAH. Our results
demonstrated no significant differences in clinically evident
or TCD-evident vasospasm between the two groups. When
regional CBF disturbances were evaluated, no group differ-
ences were observed. Follow-up examinations using GOS as-
sessments, SPECT scanning, and neuropsychological function
evaluations did not reveal any significant differences. There
were additional costs and more complications for the hyper-
dynamic treatment group. In summary, the outcome mea-
sures did not reveal any significant differences between the
two SAH treatment models.
ACKNOWLEDGMENTS
We thank the nurses of the Department of Neurosurgery
for skillful help and cooperation in performing this study.
Received, December 4, 2000.
Accepted, April 19, 2001.
Reprint requests: Bertil Romner, M.D., Ph.D., Department of Neuro-
surgery, University Hospital, S-22185 Lund, Sweden. Email:
bertil.romner@neurokir.lu.se
REFERENCES
1. Aaslid R, Markwalder TM, Nornes H: Noninvasive transcranial
Doppler ultrasound recording of flow velocity in basal cerebral
arteries. J Neurosurg 57:769–774, 1982.
2. Adams HP Jr, Kassell NF, Torner JC, Haley EC Jr: Predicting
cerebral ischemia after aneurysmal subarachnoid hemorrhage:
Influences of clinical condition, CT results, and antifibrinolytic
therapy—A report of the Cooperative Aneurysm Study. Neurol-
ogy 37:1586–1591, 1987.
3. Awad IA, Carter LP, Spetzler RF, Medina M, Williams FC Jr:
Clinical vasospasm after subarachnoid hemorrhage: Response to
hypervolemic hemodilution and arterial hypertension. Stroke 18:
365–372, 1987.
4. Beck AT, Ward CH, Mendelson M, Mock J, Erbaugh J: An inventory
for measuring depression. Arch Gen Psychiatry 4:225–236, 1961.
5. Bornstein RA, Weir BK, Petruk KC, Disney LB: Neuropsycholog-
ical function in patients after subarachnoid hemorrhage. Neuro-
surgery 21:651–654, 1987.
6. Darby JM, Yonas H, Marks EC, Durham S, Snyder RW, Nemoto
EM: Acute cerebral blood flow response to dopamine-induced
hypertension after subarachnoid hemorrhage. J Neurosurg 80:
857–864, 1994.
7. Ekelund A, Säveland H, Reinstrup P, Brandt L: Additional colloids
have only a minor haemodilutive effect after surgery for aneurysmal
subarachnoid hemorrhage. Br J Neurosurg 13:399–404, 1999.
8. Ekelund A, Säveland H, Romner B, Brandt L: Is transcranial
Doppler sonography useful in detecting late cerebral ischemia
after aneurysmal subarachnoid hemorrhage? Br J Neurosurg 10:
19–25, 1996.
Neurosurgery, Vol. 49, No. 3, September 2001
603
9. Feigin VL, Rinkel GJ, Algra A, Vermeulen M, van Gijn J: Calcium
antagonists in patients with aneurysmal subarachnoid hemor-
rhage: A systematic review. Neurology 50:876–883, 1998.
10. Fisher CM, Kistler JP, Davis JM: Relation of cerebral vasospasm to
subarachnoidal hemorrhage visualized by computerized tomo-
graphic scanning. Neurosurgery 6:1–9, 1980.
11. Fisher CM, Robertson GH, Ojemann RG: Cerebral vasospasm
with ruptured saccular aneurysms: The clinical manifestations.
Neurosurgery 1:245–248, 1977.
12. Golden CJ: Stroop Color and Word Test. Chicago, Stoelting, 1978.
13. Grosset DG, Straiton J, du Trevou M, Bullock R: Prediction of
symptomatic vasospasm after subarachnoid hemorrhage by rap-
idly increasing transcranial Doppler velocity and cerebral blood
flow changes. Stroke 23:674–679, 1992.
14. Heaton RK, Grant I, Matthews CG: Comprehensive Norms for an
Expanded Halstead-Reitan Neuropsychological Test Battery. Odessa,
Psychological Assessment Resources, 1993.
15. Heros RC, Morcos JJ: Cerebrovascular surgery: Past, present, and
future. Neurosurgery 47:1007–1033, 2000.
16. Hijdra A, van Gijn J, Nagelkerke NJ, Vermeulen M, van Crevel
H: Prediction of delayed cerebral ischemia, rebleeding, and out-
come after aneurysmal subarachnoid hemorrhage. Stroke
19:1250–1256, 1988.
17. Hunt WE, Hess RM: Surgical risk as related to time of interven-
tion in the repair of intracranial aneurysms. J Neurosurg 28:14–19,
1968.
18. Hütter BO, Gilsbach JM: Which neuropsychological deficits are
hidden behind a good outcome (Glasgow⫽1) after aneurysmal
subarachnoid hemorrhage? Neurosurgery 33:999–1006, 1993.
19. Hop JW, Rinkel GJ, Algra A, van Gijn J: Case-fatality rates and
functional outcome after subarachnoid hemorrhage: A systematic
review. Stroke 28:660–664, 1997.
20. Jennett B, Bond M: Assessment of outcome after severe brain
damage: A practical scale. Lancet 1:480–484, 1975.
21. Kapp J, Mahaley MS Jr, Odom GL: Cerebral arterial vasospasm:
Part 2—Experimental evaluation of mechanical and humoral fac-
tors in pathogenesis. J Neurosurg 29:339–349, 1968.
22. Kassell NF, Peerless SJ, Durward QJ, Beck DW, Drake CG, Adams
HP Jr: Treatment of ischemic deficits from vasospasm with intra-
vascular volume expansion and induced arterial hypertension.
Neurosurgery 11:337–343, 1982.
23. Kassell NF, Torner JC, Haley EC Jr, Jane JA, Adams HP, Kongable
GL: The International Cooperative Study on the Timing of Aneu-
rysm Surgery: Part I—Overall management results. J Neurosurg
73:18–36, 1990.
24. Keller TS, McGillicuddy JE, LaBond VA, Kindt GW: Modification
of focal cerebral ischemia by cardiac output augmentation. J Surg
Res 39:420–432, 1985.
25. Kindt GW, McGillicuddy JE, Giannotta SL, Pritz MB: The reversal
of neurologic deficit in patients with acute cerebral ischemia by
profound increases in intravascular volume, in Gotoh F, Nagai H,
Tazaki Y (eds): Cerebral Blood Flow and Metabolism. Copenhagen,
Munksgaard, 1979.
26. Kosnik EJ, Hunt WE: Postoperative hypertension in the manage-
ment of patients with intracranial aneurysms. J Neurosurg 45:
148–154, 1976.
27. Lassen NA: On the regulation of cerebral blood flow in diseases of
the brain with special regard to “luxury perfusion syndrome” of
brain tissue, i.e., a syndrome characterized by relative hyperemia or
absolute hyperemia of the brain tissue. Prog Brain Res 30:121–124,
1968.
604 Egge et al.
28. Lennihan L, Mayer SA, Fink ME, Beckford A, Paik MC, Zhang H,
Wu YC, Klebanoff LM, Raps EC, Solomon RA: Effect of hypervo-
lemic therapy on cerebral blood flow after subarachnoid hemor-
rhage: A randomized controlled trial. Stroke 31:383–391, 2000.
29. Levy ML, Giannotta SL: Cardiac performance indices during hyper-
volemic therapy for cerebral vasospasm. J Neurosurg 75:27–31, 1991.
30. Lezak MD: Neuropsychological Assessment. New York, Oxford Uni-
versity Press, 1995, ed 3.
31. Lindegaard KF, Nornes H, Bakke SJ, Sorteberg W, Nakstad P: Cere-
bral vasospasm diagnosis by means of angiography and blood ve-
locity measurements. Acta Neurochir (Wien) 100:12–24, 1989.
32. Ljunggren B, Sonesson B, Säveland H, Brandt L: Cognitive im-
pairment and adjustment in patients without neurological deficits
after aneurysmal SAH and early operation. J Neurosurg 62:673–
679, 1985.
33. Lund-Johansen M, Hugdahl K, Wester K: Cognitive function in
patients with Parkinson’s disease undergoing stereotaxic thalam-
otomy. J Neurol Neurosurg Psychiatry 60:564–571, 1996.
34. Maroon JC, Nelson PB: Hypovolemia in patients with subarach-
noid hemorrhage: Therapeutic implications. Neurosurgery 4:223–
226, 1979.
35. Masdeu J, Brass L, Holman L, Kushner M: Brain single photon
emission computed tomography. Neurology 44:1970–1977, 1994.
36. Matsui T, Asano T: The hemodynamic effects of prolonged albu-
min administration in beagle dogs exposed to experimental sub-
arachnoid hemorrhage. Neurosurgery 32:79–84, 1993.
37. Mavaddat N, Sahakian BJ, Hutchinson PJ, Kirkpatrick PJ: Cogni-
tion following subarachnoid hemorrhage from anterior commu-
nicating artery aneurysm: Relation to timing of surgery.
J Neurosurg 91:402–407, 1999.
38. Mayberg MR, Batjer HH, Dacey R, Diringer M, Haley EC, Heros
RC, Sternau LL, Torner J, Adams HP Jr, Feinberg W, Thies W:
Guidelines for the management of aneurysmal subarachnoid
hemorrhage: A statement for healthcare professionals from a
special writing group of the Stroke Council, American Heart
Association. Stroke 25:2315–2328, 1994.
39. Mayer SA, Solomon RA, Fink ME, Lennihan L, Stern L, Beckford
A, Thomas CE, Klebanoff LM: Effect of 5% albumin solution on
sodium balance and blood volume after subarachnoid hemor-
rhage. Neurosurgery 42:759–768, 1998.
40. Medlock MD, Dulebohn SC, Elwood PW: Prophylactic hypervol-
emia without calcium channel blockers in early aneurysm sur-
gery. Neurosurgery 30:12–16, 1991.
41. Miller EN, Satz P, Visscher B: Computerized and conventional
neuropsychological assessment of HIV-1-infected homosexual
men. Neurology 41:1608–1616, 1991.
42. Mizuno M, Nakajima S, Sampei T, Nishimura H, Hadeishi H,
Suzuki A, Yasui N, Nathal-Vera E: Serial transcranial Doppler
flow velocity and cerebral blood flow measurements for evalua-
tion of cerebral vasospasm after subarachnoid hemorrhage.
Neurol Med Chir (Tokyo) 34:164–171, 1994.
43. Mori K, Arai H, Nakajima K, Tajima A, Maeda M: Hemorheologi-
cal and hemodynamic analysis of hypervolemic hemodilution
therapy for cerebral vasospasm after aneurysmal subarachnoid
hemorrhage. Stroke 26:1620–1626, 1995.
44. Muizelaar JP, Becker DP: Induced hypertension for the treatment
of cerebral ischemia after subarachnoid hemorrhage: Direct effect
on cerebral blood flow. Surg Neurol 25:317–325, 1986.
45. Öhman J, Servo A, Heiskanen O: Risk factors for cerebral infarc-
tion in good-grade patients after aneurysmal subarachnoid hem-
orrhage and surgery: A prospective study. J Neurosurg 74:14–20,
1991.
Neurosurgery, Vol. 49, No. 3, September 2001
46. Olsen TS, Larsen B, Skriver EB, Herning M, Enevoldsen E, Lassen
NA: Focal cerebral hyperemia in acute stroke: Incidence, patho-
physiology and clinical significance. Stroke 12:598–607, 1981.
47. Origitano TC, Wascher TM, Reichman OH, Anderson DE: Sus-
tained increased cerebral blood flow with prophylactic hyperten-
sive hypervolemic hemodilution (“triple-H” therapy) after sub-
arachnoid hemorrhage. Neurosurgery 27:729–739, 1990.
48. Polin RS, Coenen VA, Hansen CA, Shin P, Baskaya MK, Nanda A,
Kassell NF: Efficacy of transluminal angioplasty for the manage-
ment of symptomatic cerebral vasospasm following aneurysmal
subarachnoid hemorrhage. J Neurosurg 92:284–290, 2000.
49. Powers WJ, Grubb RL, Baker RP, Mintun MA, Raichle ME: Re-
gional cerebral blood flow and metabolism in reversible ischemia
due to vasospasm. J Neurosurg 62:539–546, 1985.
50. Pritz MB: Treatment of cerebral vasospasm due to aneurysmal
hemorrhage: Past, present and future of hyperdynamic therapy.
Neurosurg Q 7:273–285, 1997.
51. Pritz MB, Giannotta SL, Kindt GW, McGillicuddy JE, Prager RL:
Treatment of patients with neurologic deficits associated with
cerebral vasospasm by intravascular volume expansion. Neuro-
surgery 3:364–368, 1978.
52. Reitan RM, Wolfson D: The Halstead-Reitan Neuropsychological Test
Battery: Theory and Clinical Interpretation. Tucson, Neuropsychol-
ogy Press, 1993, ed 2.
53. Rosenørn J: Unfavorable prognostic factors in patients with intra-
cranial aneurysms and the possibilities to improve the overall
outcome. Prog Clin Neurosci 2:179–186, 1988.
54. Säveland H, Brandt L: Which are the major determinants for
outcome in aneurysmal subarachnoid hemorrhage? Acta Neurol
Scand 90:245–250, 1994.
55. Scandinavian Stroke Study Group: Multicenter trial of hemodilu-
tion in ischemic stroke: Background and study protocol. Stroke
16:885–890, 1985.
56. Shimoda M, Oda S, Tsugane R, Sato O: Intracranial complications
of hypervolemic therapy in patients with a delayed ischemic
deficit attributed to vasospasm. J Neurosurg 78:423–429, 1993.
57. Solomon RA, Fink ME, Lennihan L: Early aneurysm surgery and
prophylactic hypervolemic hypertensive therapy for the treat-
ment of aneurysmal subarachnoid hemorrhage. Neurosurgery
23:699–704, 1988.
58. Solomon RA, Fink ME, Lennihan L: Prophylactic volume expan-
sion therapy for the prevention of delayed cerebral ischemia after
early aneurysm surgery. Arch Neurol 45:325–332, 1988.
59. Solomon RA, Post KD, McMurtry JG: Depression of circulating
blood volume after subarachnoid hemorrhage: Implications for
treatment of symptomatic vasospasm. Neurosurgery 15:354–361,
1994.
60. Spreen O, Strauss E: A Compendium of Neuropsychological Tests.
New York, Oxford University Press, 1998, ed 2.
61. Stroop JR: Studies of interference in serial verbal reactions. J Exp
Psychol 18:643–662, 1935.
62. Tranmer BI, Keller TS, Kindt GW, Archer D: Loss of cerebral
regulation during cardiac output variations in focal cerebral isch-
emia. J Neurosurg 77:253–259, 1992.
63. Trumble ER, Muizelaar JP, Myseros JS, Choi SC, Warren BB:
Coagulopathy with the use of hetastarch in the treatment of
vasospasm. J Neurosurg 82:44–47, 1995.
64. Vora YY, Suarez-Almazor M, Steinke DE, Martin ML, Findlay JM:
Role of transcranial Doppler monitoring in the diagnosis of cere-
bral vasospasm after subarachnoid hemorrhage. Neurosurgery
44:1237–1248, 1999.
 Hyperdynamic Postoperative Fluid Therapy after SAH
65. Waldemar G, Hasselbalch SG, Andersen AR, Delecluse F,
Petersen P, Johnsen A, Paulsen OB: 99mTc-d,l-HMPAO and SPECT
of the brain in normal aging. J Cereb Blood Flow Metab 11:508–
521, 1991.
66. Wechsler D: Wechsler Adult Intelligence Scale. New York, The Psy-
chological Corporation, 1955.
67. Wechsler D: Wechsler Memory Scale-Revised Manual. New York,
The Psychological Corporation, 1987.
68. Wetzel JW: Clinical Handbook of Depression. New York, Gardner, 1984.
69. Wijdicks EF, Vermeulen M, Hijdra A, van Gijn J: Hyponatremia
and cerebral infarction in patients with ruptured intracranial
aneurysms: Is fluid restriction harmful? Ann Neurol 17:137–140,
1985.
70. Yamakami I, Isobe K, Yamaura A: Effects of intravascular
volume expansion on cerebral blood flow in patients with
ruptured cerebral aneurysms. Neurosurgery 21:303–309, 1987.
COMMENTS
This prospective, randomized study compares the efficacy
of early induced hypertensive hypervolemic hemodilution
combination therapy (triple-H therapy) in a group of patients
with aneurysmal subarachnoid hemorrhage (SAH) with a
control group with normovolemic, normotensive treatment
and normal hematocrit after aneurysmal SAH. All of the
patients received intravenous nimodipine. Each group com-
prised 16 patients, and the treatment period lasted until Day
12 after the SAH. The stratification was according to Fisher
grade, and the other important prognostic measures such as
Hunt and Hess grade at arrival, the patient’s age, and the
location of the aneurysm were evenly distributed. The pa-
tients who received triple-H therapy showed no difference
in outcome at the time of discharge or 1 year after therapy.
Other outcome measures, clinically and transcranial Doppler
sonography-detected vasospasm, early or late single-photon
emission computed tomography findings, and neuropsycho-
logical variables did not show any differences either. There
were more complications in the active treatment group, and
this care was more expensive, even without taking into ac-
count the heavy nursing protocol. The authors conclude that
prophylactic triple-H treatment does not improve patient out-
come, does not prevent vasospasm, and may be harmful. The
message is clear, and no comparable study has been done
before. However, this study does not answer the question
whether the treatment has some effect on symptomatic clini-
cal vasospasm, where it is actually recommended.
The study is elaborate. The authors studied 32 patients,
who were evenly distributed between symptomatic and con-
trol groups. The caseload in the department was only 12 to 13
patients per year. There were no patients with Hunt and Hess
Grades 4 and 5 in the series; these patients were excluded
because of late operations. The incidence of vasospasm was
high at 70%, but in general the patient outcomes in the series
were quite good.
The single-photon emission computed tomographic study
did not provide much information, and comparison of studies
taken from both sides was used as a measure of cerebral blood
flow reduction. This methodology is suitable for patients with
middle cerebral artery aneurysms, but there were other loca-
tions with more global changes. Late single-photon emission
Neurosurgery, Vol. 49, No. 3, September 2001
605
computed tomography was not shown to be a reliable control
method for infarcts after delayed vasospasm. The use of late
computed tomographic or magnetic resonance imaging veri-
fication perhaps could be a better measure for these patients.
The neuropsychological studies that the authors used did not
offer much information either, but it is a well-known fact that
even in larger series of patients with SAH, the differences are
difficult to find.
Timo Koivisto
Matti Vapalahti
Kuopio, Finland
In this small, prospective series of patients with aneurys-
mal SAH, prophylactic triple-H therapy was advantageous in
terms of reducing the clinical or transcranial Doppler sonog-
raphy incidence of cerebral vasospasm. Prophylactic triple-H
therapy also did not have a beneficial effect on long-term
clinical outcome in these patients. Not surprisingly, it was
associated with higher costs and more complications. The
major deficiency in this study is the very small number of
patients studied. It is possible that triple-H therapy had a
small benefit that was not demonstrated or that was obscured
by the small sample size. Caution must be exercised not to
extrapolate the lack of efficacy of prophylactic triple-H therapy
on the outcomes of patients with aneurysmal SAH in this study
to therapeutic triple-H therapy, for which there are a much
stronger theoretical basis and more robust anecdotal data.
Neal F. Kassell
Charlottesville, Virginia
In this interesting study, the authors conclude that prophy-
lactic triple-H therapy does not lead to improved patient out-
come after aneurysmal SAH but is associated with a higher
complication rate and increased costs. The authors recognize
that there were only a few patients in this study and that there is
considerable chance of ␤-type error. Nevertheless the results of
this study are rather convincing. Why, then, are the results of
management of SAH thought to be better now than they were 20
years ago? Neurosurgeons assume that vasospasm leads to di-
minished blood flow and oxygen delivery to the brain. Treat-
ment or prophylaxis of this problem with hypervolemia makes
no sense, because blood volume and cardiac output are not
represented in the Poiseuille viscosity coefficient, and manipu-
lation of these factors does not increase cerebral blood flow or
oxygen delivery. However, hypervolemia is still used, because
with hypovolemia, it would be difficult to induce arterial hyper-
tension if that became necessary. In addition, it might prevent
blood pressure decreases—short-lived but deleterious nonethe-
less—associated with the use of nimodipine. In this respect, one
should keep in mind that slow, continuous intravenous admin-
istration of nimodipine (as used in this study) is not available in
the United States. Thus, in the United States, a certain degree of
hypervolemia might be more beneficial than it would be in
Europe, but it is certainly not worth provoking cardiac failure
and pulmonary edema.
As for hemodilution, I note that in the treatment group,
hematocrit was well below my ideal of 33 to 35, whereas in the
606 Egge et al.

control group, it was exactly in this range. It is at a hematocrit
level of approximately 34 that a further reduction in hemat-
ocrit does not lead to any significant further reduction in
blood viscosity (i.e., no further effect on blood flow according
to the Poiseuille viscosity coefficient), but the decrease in
oxygen carrier capacity easily amounts to 10 to 15%. Hence, I
transfuse patients who have a hematocrit level less than 31
and are within the time window for vasospasm (which would
apply here to the treatment group from Days 7 through 11).
Finally, with regard to hypertension, I do not believe that
many clinicians advocate the use of prophylactic hyperten-
sion, partly because of the development of tachyphylaxis for
␣-adrenergic drugs by the time they become really crucial
(i.e., the appearance of delayed ischemic deficits). Neverthe-
less, many of my patients are administered these agents in
low doses early on, not to induce hypertension but to treat
hypotensive episodes.
In summary, this thought-provoking study in no way
proves that the judicious use of prophylactic hypervolemia,
hemodilution to a hematocrit level of 34, and use of hyper-
tension only in case of delayed ischemic deficits have led to
better outcomes for patients today as compared with 20 years
ago. This article is not sufficient reason to abandon these
treatment modalities at this time.
J. Paul Muizelaar
Sacramento, California
The authors prospectively studied two patient cohorts after
surgical treatment of aneurysmal SAH. One patient group
received prophylactic triple-H therapy and was compared
with a group of patients who were administered normovol-
emic fluid therapy with normotension. The authors found no
significant differences between the two groups according to a
variety of short- and long-term outcome measures.
This study is rated Level of Evidence II (data from random-
ized trials with high false-positive [␣] and high false-negative
errors [␤]) according to American Heart Association criteria
(1, 2). It has been comprehensively analyzed. Unfortunately,
as with any single-institution clinical study, its statistical
power is limited because of the small sample size, and the
finding of no difference between study groups, although sug-
gestive, remains unproved. Nevertheless, the American Heart
Association’s evidence-based review (2) of the treatment of
patients with SAH did not identify any controlled clinical
trials that demonstrated the efficacy of triple-H therapy but
did recommend the use of this therapy on the basis of uncon-
trolled studies. This report from Egge et al. suggests that the
primary effect of triple-H therapy is to increase morbidity and
cost, which should prompt a critical reappraisal of this treat-
ment modality.
Christopher L. Taylor
Warren R. Selman
Cleveland, Ohio
1. Cook DJ, Guyatt GH, Laupacis A, Sackett DL: Rules of evidence
and clinical recommendations on the use of antithrombotic agents.
Chest 102[4 Suppl]:305S–311S, 1992.
2. Mayberg MR, Batjer HH, Dacey R, Diringer M, Haley EC, Heros
RC, Sternau LL, Torner J, Adams HP Jr, Feinberg W: Guidelines for
the management of aneurysmal subarachnoid hemorrhage. A
statement for healthcare professionals from a special writing group
of the Stroke Council, American Heart Association. Stroke 25:
2315–2328, 1994.
Although carefully designed, this study does not answer
the question whether prophylactic triple-H therapy after an-
eurysmal SAH is efficacious for reducing deficits related to
cerebral vasospasm. In modern series with calcium channel
blockers, the incidence of neurological deficits from vaso-
spasm in patients with Hunt and Hess Grades I through III
lesions (the same as patients enrolled in this study) is approx-
imately 15% or less (1). Thus, a poor outcome because of
vasospasm might have been expected in only 2 of the 16
controls used. Obviously, these numbers are too small for any
meaningful comparison. This study should not be considered
as proof for or against the use of triple-H therapy, but rather
should be viewed as a template for the design of an ade-
quately powered, prospective, randomized trial.
Marc R. Mayberg
Cleveland, Ohio
1. Mayberg MR, Batjer HH, Dacey R, Diringer M, Haley EC, Heros
RC, Sternau LL, Torner J, Adams HP Jr, Feinberg W: Guidelines for
the management of aneurysmal subarachnoid hemorrhage. A
statement for healthcare professionals from a special writing group
of the Stroke Council, American Heart Association. Stroke 25:
2315–2328, 1994.

Congress of Neurological Surgeons’

Mission Statement
“The Congress of Neurological Surgeons exists for the purpose of promoting the public
welfare through the advancement of neurosurgery, by a commitment to excellence in
education, and by dedication to research and scientific knowledge. The Congress of
Neurological Surgeons maintains the vitality of our learned profession through the
altruistic volunteer efforts of our members and the development of leadership in service
to the public, to our colleagues in other disciplines, and to the special needs of our fellow
neurosurgeons throughout the world and at every stage of their professional lives.”

Neurosurgery, Vol. 49, No. 3, September 2001