Acta Neurochir (Wien) (1999) 141: 1069±1073
Acta Neurochirurgica
> Springer-Verlag 1999
Printed in Austria
Circulating Blood Volume in Patients with Subarachnoid Haemorrhage
K. Sato1, H. Karibe2, and T. Yoshimoto2
1 Department of Neuroanesthesia, Kohnan Hospital, Japan
2 Department of Neurosurgery, Tohoku University School of Medicine, Japan
Summary
To establish the ¯uid management for patients with subarachnoid
haemorrhage (SAH), circulating blood volume (BV) was measured
by pulse-spectrophotometry using indocyanine green (ICG) in 34
cases with SAH and 20 cases with neurosurgical disorders as control.
BV measured immediately after induction of anaesthesia was lower
in cases with SAH than that in controls. (62.8 G 12.3 vs.
73.3 G 11.2 ml/kg, p < 0:01). In cases with SAH, the BV was signi®cantly decreased in females (p < 0:05) but not so signi®cantly in
males. In female cases with SAH, reduced BV was increased 3 days
after operation (p < 0:01). In conclusion BV is decreased in cases
with SAH, especially in females. Active ¯uid therapy may be necessary when temporary vascular occlusion is required during aneurysm
surgery. Since hypovolaemia may cause symptomatic vasospasm,
BV measurement with pulse-spectrophotometry may provide useful
information to insure normovolaemia.
Keywords: Blood volume; subarachnoid haemorrhage; pulsespectrophotometry; ICG; spasm.
Introduction
It has been reported that circulating blood volume
(BV) is reduced in cases with subarachnoid haemorrhage (SAH) [11], and that hypovolaemia worsens cerebral ischaemia resulting from delayed cerebral vasospasm [16]. Temporary vessel occlusion has been
widely used during early surgery for ruptured cerebral
aneurysm [8] under employment of brain protection to
tolerate cerebral ischaemia induced by temporary vessel occlusion [10]. Since hypovolaemia may decrease
collateral circulation and worsen cerebral ischaemia
during temporary vessel occlusion, it is important to
establish ¯uid management for patients who receive
early surgery for SAH.
It has been di½cult to measure BV in an operating
room since radio-isotopes have been used to measure
BV [18]. Recently developed pulse-spectrophotometry
using indocyanine green (ICG) is a less-invasive and
simple technique for measuring BV. In this study, to
establish ¯uid management in patients with subarachnoid haemorrhage, BV was measured in neurosurgical patients using recently developed pulse-spectrophotometry techniques.
Patients and Methods
A total of 54 cases including 34 with SAH (female 25, male 9; age
range 35 to 87, mean age 57.1) and 20 with other neurosurgical disorders (non-SAH) (female 8, male 12; age range 34 to 74, mean age
60.5), underwent the following protocol with approval of the ethics
committee of Kohnan Hospital and informed consent. Cases with
SAH were operated on within 72 hours from the onset and pre-operative Hunt and Kosnik's grades were 1 in 8 cases, 2 in 16, 3 in 12, 4
in 4 cases. Disorders in the control patients were unruptured cerebral
aneurysms, intracerebral haemorrhage, meningioma, pituitary adenoma, cervical spondylosis, cerebral infarction in 8, 8, 4, 2, 2, 1
cases respectively.
Anaesthesia was induced by propofol (1±2 mg/kg) and fentanyl
(5±10 mg/kg). The trachea was intubated after muscle relaxation by
0.08 mg/kg of vecuronium bromide. After induction of anaesthesia,
a central venous catheter was inserted from the right internal jugular
vein. After achievement of stabilized systemic circulation, 20 mg of
ICG was administrated through the central venous catheter. Using
dye densitogram analyzing system (DDG-2001, Nihon Kohden Inc.,
Tokyo, Japan), circulating blood volume (BV) was measured in all
cases. In cases with SAH, BV was also measured immediately, 1-, 3-,
and 7 days afer operation. Normovolaemic management was achieved by assessing circulation and urine amount during anaesthesia
and by assessing body weight, central venous pressure and sodium
balance during the post-operative period.
All parameters were expressed as mean G standard deviation.
Unpaired t-test was used to compare the two groups, SAH and nonSAH and patients' distrubution of sex were examined using chisquare test for independence. A probability value < 0:05 was considered signi®cant.
Results
Female cases were more frequent in SAH than nonSAH (p < 0:05); however there was no statistically
1070
Fig. 1. BV in female patients after the induction of anesthesia. BV in
female cases with SAH was lower than non-SAH female patient
Fig. 2. BV in male cases after the induction of anesthesia.. BV was
not decreased in male patients with SAH
signi®cant di¨erence in the distribution of age between
SAH and non-SAH groups.
BV after induction of anaesthesia was signi®cantly
lower in SAH than non-SAH (62.8 G 12.3 vs.
73.3 G 11.2 mg/kg, p < 0:01). BV was signi®cant
lower in SAH than non-SAH in females (56.7 G 10.1
K. Sato et al.
Fig. 3. Change of BV in women patients. 3 days after operation, BV
increased signi®cantly
vs. 70.3 G 13.8 ml/kg, p < 0:01, Fig. 1). However BV
was not di¨erent between the two groups in male
(71.5 G 14.3 vs. 75.4 G 9.2 ml/kg, Fig. 2).
In female cases with SAH, BV immediately, 1-, 3-,
and 7-days after surgery were 62.1 G 14.1, 61.0 G
11.2, 73.4 G 16.3, and 76.6 G 18.5 ml/kg respectively.
BV at 3- and 7-days after surgery was signi®cantly
higher than that after the induction of anaesthesia
(p < 0:01, Fig. 3).
There was no signi®cant correlation between age
and BV in cases with SAH (rs ˆ ÿ0:0982) nor between
Hunt and Kosnik's aneurysmal grade (rs ˆ ÿ0:227).
Symptomatic cerebral vasospasm (SVS) was observed in 8 cases with SAH. There was no signi®cant
di¨erence in BV between cases with and without
SVS after induction of anaesthesia (65.0 G 15.2 vs.
62.1 G 11.5 ml/kg).
In 3 female cases with angiographlcally con®rmed
SVS, change in BV was followed (Fig. 4). In each case
pre-operative Hunt and kosnik grade was 3. In case 1
(77 y.o., anterior communicating artery aneurysm),
BV increased immediately then declined 7 days after
operation corresponding to the occurrence of SVS. In
case 2 (53 y.o., right middle cerebral artery aneurysm),
BV remained low throughout the postoperative period, and then outbreak of SVS was seen. In case 3 (63
y.o., left middle cerebral artery aneurysm), BV in-
Circulating Blood Volume in Patients with Subarachnoid Haemorrhage
Fig. 4. Change of BV in cases with SVS. In case 1, SVS developed
corresponding to decline of BV 7 days after operation. In case 2, BV
remained low throughout the postoperative period and SVS developed 10 days after operation. In case 3, BV increased postoperatively
to within normal range, although SVS occurred 3 days after surgery
5 case 1, 4 case 2, case 3
creased postoperatively to remain within normal
range, but SVS occurred 3 days after operation.
Discussion
As standard technique for the measurement of circulating blood volume (BV), dilution method using
radio-isotope has been established [18]. However it has
not been generally applied in clinical situations because of continuous radiation by the remaining radioactive tracer in the body. A measuring method using
tricarbocyanine dye, indocyanine green (ICG) has
been introduced. However, it was not widely used,
since intermittent or continuous sampling and its
measurement were required to obtain a time-density
curve of ICG. Recently developed pulse-spectrophotometry technique has enabled one to measure arterial
blood haemoglobin/ICG ratio less invasively and easily, and to calculate BV [6]. Reproducibility of this
technique has been established already [5, 7]. In this
study, using this pulse-spectrophotometry technique,
BV was measured in neurosurgical cases. The present
study demonstrates that BV is measurable in most
cases under general anaesthesia. However, in a few
cases with depressed peripheral circulation, it could
not be achieved as well as oxygen saturation measurement using a pulse-oximeter. In addition, BV has not
1071
always been able to be followed postoperatively, because BV measurement has not been possible in patients who su¨ered restlessness.
It has been known that BV decreased in patients
with SAH. Using red blood cell ( 51 Cr-RBC) and human serum albumin (HAS), labelled with radio-isotope, Maroon et al. measured BV in 15 cases with SAH
and in 6 with other neurosurgical disorders. They reported that BV was more signi®cantly reduced in cases
with SAH that in others [11]. Solomon et al also demonstrated BV reduction in 15 female cases with SAH
[16]. Nelson et al. reported the relationship between
SAH appearance on CT and BV measured with 125 IHAS. In all cases with reduced BV, the basal cisterns
have been compressed or obliterated on the CT scans.
In contrast, in cases with normal BV, such ®ndings on
CT were recognized only in 12.5% of the cases. They
demonstrated higher incidence of depressed BV when
elevated intracranial pressure was suggested on CT
[14]. The present study clearly showed depressed BV in
female cases with SAH, as previously described [16]
and the present results suggest the reliability of pulsespectrophotometry using ICG. The reason is not clear
why the BV was depressed only in female cases. Further study will be required to clarify the mechanism by
measuring various hormones.
Delayed cerebral vasospasm is an important factor
for determining morbidity and mortality after SAH
[17]. It has been suggested that hypovolaemia after
SAH might play an important role in the occurrence of
symptomatic vasospasm (SVS) [11]. In fact, it has been
demonstrated that BV is signi®cantly decreased in
cases with SVS [16]. Hypervomlaemic haemodilution
therapy has been widely used for the management of
SVS [4, 20]. However, it has also been reported that
volume overload rather than normovolaemia is ineffective [22]. In the present study, BV recovered to the
normal range within 3 days after surgery even in cases
whose pre-operative BV has been depressed. Therefore, BV is considered to normalize at the period when
SVS usually occur in most cases. In this study, the time
courses of BV changes were followed in 3 cases with
SVS. Recovered BV decreases at the time when SVS
occurred in case 1, and depressed BV did not recover
and SVS broke out in case 2. In contrast, BV remained
within normal range in case 3. These results suggest
that depressed BV may take a part in the occurrence of
SVS, however, only maintenance of BV within normal
range may not be enough to avoid SVS.
Several factors have been proposed as mechanisms
1072
of BV reduction in patients with SAH: bed ridden, supine position, negative daily balance of nitrogen, reduced red blood cell production, iatrogenic blood loss
[11]. In the present study, however, all cases received
surgery within 72 hours after onset of SAH. Thus it is
unlikely that BV is reduced by the mechanisms mentioned above in such a short period. As another mechanism of hypovolaemia after SAH, it has been reported that sodium-dependent diuresis is induced by
the increased production of natriuretic peptide system
(ANP, BNP, CNP) after SAH [2, 9] to result in BV
reduction [21]. However, this sodium-dependent diuresis induced by natriuretic peptide system occurs 5±7
days after SAH [13, 19], it may not account for BV reduction immediately after SAH. Sympathetic hyperactivity may account for hypovolaemia immediately
after SAH. It has been reported that blood concentration of catecholamines increased in cases with SAH [1].
It has also been demonstrated that BV decreases under
conditions of increased sympathetic activity [12].
These results suggest that catecholamine storm immediately after SAH reduces systemic vascular compliance resulting in BV reduction.
Temporary vessel occlusion is widely used during
cerebral aneurysm surgery [8]. Brain protection is required against ischaemia during vessel occlusion [10].
Hypovolaemia should be avoided, since it may worsen
collateral circulation during temporary vessel occlusion. The present results suggest that hypovolaemia
often occurs immediately after induction of anaesthesia especially in female cases with SAH. Therefore,
aggressive ¯uid management is necessary in such cases.
Generally 500±1000 ml of crystaloid is administrated
during induction of anaesthesia, however it is not
enough in some cases with SAH. Alpha stimulant is
sometimes required to normalize hypotension immediately after induction of anaesthesia, even in cases
whose pre-operative blood pressure is barely controlled using vasodilators in practice. Cerebral perfusion pressure should be maintained using vasomotor
agents, and ¯uid management to maintain normo- or
slight hyper-volaemia should be required before temporary vascular occlusion. In addition, abnormality in
blood-brain barrier and/or in autoregulation may exist
in cases with SAH, in whom overhydration may lead
to aggravation of cerebral oedema. In such cases, strict
¯uid management would be necessary by pulmonary
artery catheter.
In conclusion, hypovolaemia may occur in cases
with SAH, especially in women. Attention should be
K. Sato et al.
paid when temporary vessel occlusion is required during early surgery. Postoperatively, symptomatic vasospasm could occur corresponding to reduction of circulating blood volume. Assessment of body weight,
central venous pressure and daily sodium balance are
important but su½cient for estimation of circulating
blood volume. For optimal ¯uid management for
patients with subarachnoid haemorrhage, pulsespectrophotometry may be useful for assessing blood
volume.
References
1. Benedict CR, Loach AB (1978) Sympathetic nervous system
activity in patients with subarachnoid hemorrhage. Stroke 9:
237±244
2. Berendes E, Walter M, Cullen P, Prien T, Van Aken H, Horsthemke J, Schulte M, von Wild K, Scherer R (1997) Secretion of
brain natriuretic peptide in patients with aneurysmal subarachnoid haemorrhage. Lancet 349, 245±249
3. Bradley EC, Barr JW (1968) Determination of blood volume
using indocyanine green (cardio-green) dye. Life Sci 7: 1001±
1007
4. Findlay JM (1997) Current management of aneurysmal subarachnoid hemorrhage guidelines from the Canadian Neurosurgical Society. Can J Neurol Sci 24: 161±170
5. Haruna M, Kumon K, Yahagi N, Watanabe Y, Ishida Y, Kobayashi M, Aoyagi T (1998) Blood volume measurement at the
bedside using ICG pulse spectrophotometry. Anesthesiology 89:
1322±1328
6. He YL, Tanigami H, Ueyama H, Mashimo T, Yoshiya I (1998)
Measurement of blood volume using indocyanine green measured with pulse-spectrophotometry: its reproducibility and reliability. Crit Care Med 26: 1446±1451
7. Iijima T, Iwao Y, Sankawa H (1998) Circulation blood volume
measured by pulse dye-densitometry. Anesthesiology 89: 1329±
1335
8. Jabre A, Symon L (1987) Temporary vascular occlusion during
aneurysm surgery. Surg Neurol 27: 47±63
9. Kurokawa Y, Uede T, Ishiguro M, Honda O, Honmou O, Kato
T, Wanibuchi M (1996) Pathogenesis of hyponatremia following
subarachnoid hemorrhage due to ruptured cerebral aneurysm.
Surg Neurol 46: 500±5007
10. Lavine SD, Masri LS, Levy ML, Giannotta SL (1997) Temporary occlusion of the middle cerebral artery in intracranial
aneurysm surgery: time limitation and advantage of brain protection. J Neurosurg 87: 817±824
11. Maroon JC, Nelson PB (1979) Hypovolemia in patients with
subarachnoid hemorrhage: therapeutic implications. Neurosurgery 4: 223±226
12. Mason DT, Bartter FC (1968) Autonomic regulation of blood
volume. Anesthesiology 29: 681±92
13. Morinage K, Hayashi S, Matsumoto Y, Omiya N, Mikami J,
Ueda M, Sato H, Inoue Y, Okawara S (1991) Serum ANP and
ADH after subarachnoid hemorrhage and hyponatremia. No
To Shinkei 43: 169±73
14. Nelson RJ, Roberts J, Rubin C, Walker V, Ackery DM, Pickard
JD (1991) Association of hypovolemia after subarachnoid hemorrhage with computed tomographic scan evidence of raised intracranial pressure. Neurosurg 29: 178±182
Circulating Blood Volume in Patients with Subarachnoid Haemorrhage
15. Ogasawara K, Kinouchi H, Nagamine Y, Koshu K, Fujiwara S,
Mizoi K (1996) Sodium balance and symptomatic vasospasm in
patients with subarachnoid hemorrhage. Surg Cereb Stroke
(Jpn) 24: 215±220
16. Solomon RA, Post KD, McMurtry JG (1984) Depression of
circulation blood volume in patients after subarachnoid hemorrhage: implications for the management of symptomatic vasospasm. Neurosurgery 15: 354±361
17. Tamargo RJ, Walter KA, Oshiro EM (1997) Aneurysmal subarachnoid hemorrhage: prognostic features and outcomes. New
Horiz 5: 364±375
18. The International Committee for Standardization in hematology (1973) Standard techniques for the measurement of red-cell
and plasma volume. Br J Haematol 25: 801±814
19. Tomida M, Muraki M, Uemura K, Yamasaki K (1998) Plasma
concentrations of brain natriuretic peptide in patients with subarachnoid hemorrhage. Stroke 29: 1584±1587
20. Weir B (1990) The history of cerebral vasospasm. Neurosurg
Clin N Am 1: 265±276
21. Wijdicks EFM, Vermeulen M, DPharm JA, Hijdra A, Bakker
WH, Gijn J (1985) Volume depletion and natriuresis in patients
with a ruptured intracranial aneurysm. Ann Neurol 18: 211±216
22. Yamakami I, Isobe K, Yamaura A (1987) E¨ects of intravascular volume expansion on cerebral blood ¯ow in patients
with ruptured cerebral aneurysms. Neurosurgery 21: 303±309
Comments
Hypovolaemia is a common ®nding in patients with subarachnoid
haemorrhage (SAH), which aggravates cerebral ischaemia. Avoiding hypovolaemia and maintaining a normovolaemic or moderate
hypervolaemic state is an essential element of contemporary treat-
1073
ment protocols for patients with SAH. Unfortunately, diagnosis of
hypovolaemia may be di½cult in the early stage after the acute haemorrhage. Moreover, during early surgery hypovolaemia may intensify ischaemic events which may occur during temporary clipping, aneurysm rupture and brain retraction.
The authors have investigated the blood volume of 34 patients
with subarachnoid haemorrhage and 20 control patients using pulsespectrophotometry based indiocyanine green technique. They found
decreased blood volume in female SAH patients compared to controls at the day of surgery, however, there was no di¨erence between
male SAH patients and controls at the day of surgery.
V. Seifert
The authors measured by pulse-spectrophotometry using indocyanine green the circulating blood volume in patients with subarachnoid haemorrhage and found a lower blood volume, especially
in female patients, compared to a control group with other neurosurgical disorders. The authors hypothesize that sympathetic hyperactivity may account for the hypovolaemia immediatly after SAH.
In my opinion, the increased intracranial pressure by the cisternal
blood with consecutive brain compression is also an important factor. The authors suggest that ¯uid management in patients with
subarachnoid haemorrhage may be improved by assessment of the
blood volume with pulse-spectrophotometry.
Today the routine postoperative management of patients with
subarachnoid hemorrhage includes already hypervolaemia and strict
volume control by central venous pressure or pulmonary wedge
pressure measurements.
R. Seiler
Correspondence: Kiyotaka Sato, Department of Neuroanesthesia,
Kohnan Hospital, 4-20-1 Nagamachi-minami, Sendai, 982-8523
Japan.