вход по аккаунту


Cerebral blood flow following normavolemic hemodilution in patients with high hematocrit.

код для вставкиСкачать
Cerebral Blood Flow
Following Normovolemic Hemodrution
in Patients with High Hematocrit
Leif H e n r i k s e n , M D , Olaf B . Paulson, M D , a n d Ronald J. Smith, MD
The effects on cerebral hemodynarnics of venisection and a 4% albumin-saline infusion were studied i n six patients
w i t h h i g h hematocrit (mean, 51.5%). Cerebral blood flow (CBF) was measured using the xenon 133 intracarotid
injection method. Blood gases were measured in arterial and jugular venous blood. Rapid two-stage hemodilution,
which lowered mean hematocrit by 9 and 1396, resulted i n CBF increases of 19 a n d 23%, respectively. J u g u l a r
venous partial pressure of oxygen and oxygen delivery capacity (CBF x arterial oxygen content) did not change
significantly from baseline. The cerebral metabolic rate for oxygen increased slightly following stage 1 hemodilution but returned to baseline value following stage 2. The study lends no support to the concept that patients
whose hematocrit is at the high end of the normal range have generalized cerebral hypoxia.
Henriksen L, Paulson OB, Smith RJ: Cerebral blood flow following normovolemic hernodilution in patients
with high hematocrit. Ann Neurol 9:454-457, 1981
T h o m a s and colleagues [ 11, 121 recently reported
that in patients with high hematocrit, hemodilution
increased cerebral blood flow ( C B F ) markedly, m o r e
than would correspond to t h e decrease i n hematocrit
and oxygen binding capacity. T h e y suggested t h a t a
state of generalized hypoxia may accompany elevated
hematocrit and that a large g r o u p of patients with
high-normal hematocrit or polycythemia might
benefit from venisection, as cerebral oxygenation
would be improved. H o w e v e r , i n a n earlier study [ 7 ] ,
w e f o u n d that hemodilution in patients with n o r m a l
hematocrit resulted in an increase which matched
only the decrease i n hematocrit and oxygen binding
capacity. The p r e s e n t study was performed to evaluate f u r t h e r t h e regulation of CBF following hemodilution in patients with high-normal hematocrit.
Material and Methods
Measurements of CBF, blood gases, and hematocrit were
made prior to and following venisection and two-stage
hemodilution in six patients whose hematocrit was high
( 5 1.5%; range, 48 to 54%), informed consent having been
obtained. The patients, 4 men and 2 women with a mean
age of 55.7 years (range, 42 to 65 years), were Caucasian
and lived at sea level. They had no signs of polycythemia or
systemic diseases that could be responsible for the elevated
hematocrit, and none had major cerebral lesions which
might interfere with autoregulation of CBF. T h e diagnoses
were transient ischemic attacks, dementia, and suspected
focal neoplastic lesions (which were not confirmed).
From the Department of Neurology, Rigshospitalet, Blegdamsvej
9, DK-2 100 Copenhagen, Denmark.
Using the Seldinger technique with percutaneous
punctures in the neck made under local anesthesia, a small
polyethylene catheter (external diameter, I .2 mm) was introduced into the internal carotid artery on the side of
the angiogram, and a second catheter (external diameter, 1.7 mm) into the contralateral internal jugular vein.
The tip of the venous catheter was placed in the superior
bulb of the internal jugular vein, and that of the arterial
catheter, in the internal carotid artery just below the
siphon. Correct positioning of the arterial catheter was
verified by noting the absence of diffuse facial discoloration
after a rapid injection of saline solution o r Evans blue.
The catheters were used to collect arterial and venous
blood for blood gas analysis. Heparinized arterial and venous blood samples were analyzed immediately after
withdrawal for pH, oxygen tension (PO,), and carbon dioxide tension (Pco2)on an ABL 1 acid-base laboratory
(Radiometer). Percent oxygen saturation was determined
spectrophotometrically. Arterial oxygen content (TaoJ
was calculated from the arterial oxygen saturation and the
hemoglobin concentration. The oxyhemoglobin dissociation curve was evaluated by calculating the oxygen tension
at 50% oxygen saturation of the blood (P5,J [ 7 ] . CBF
was measured using the xenon 133 intraarterial injection
method [6]:2 to 3 mCi of ‘“‘Xe dissolved in 2 to 3 ml of
isotonic saline solution was injected rapidly (1 to 2 sec) into
the internal carotid artery through the indwelling catheter.
Sixteen small scintillation detectors covering the lateral aspect of the hemisphere measured the clearance of the
isotope. The average clearance from these detectors (average hemispheric blood flow) was considered in the present
study. The cerebral blood flow was calculated from the iniReceived June 7, 1980. and in revised form Aug 14. Accepted for
publication Sept 22, 1080.
Address reprint requests to Dr Henriksen.
4 5 4 0364-5 134/81/050454-04$01.25 @ 1980 by the American Neurological Association
tial slope of the semilogarithmically recorded clearance
curves, an approximation of flow in the gray matter 161.
All flow values were corrected for remaining activity from
previous l”Xe injections. Corrections were also applied to
account for alterations in the blood-to-tissue partition
coefficient of l”:’Xe as a result of the hematocrit changes
Following duplicate measurements of CBF, blood gases,
and hematocrit, a two-step hemodilution was performed.
First, 400 ml of blood was withdrawn from a cubital vein
and rapidly replaced with 600 ml of saline containing 2 0 gm
of human albumin. Ten minutes after the infusion the measurements were repeated (duplicate measurements, 12minute interval). For the second step, 600 ml of saline
containing 20 gm of human albumin was rapidly infused,
and 10 minutes later the final CBF, blood gas, and
hematocrit measurements were performed (duplicate measurements, 12-minute interval). T h e mean of the duplicate
measurements was used for the calculation. Statistical
analysis was performed using Friedman’s and Wilcoxon’s
nonparametric two-way analysis of variance; p < 0.05 was
taken as the significance level. Mean values t SD of the
changes are used in the table and the figures.
Individual CBF responses in the six patients are
shown in Figure 1. The rapid two-stage hernodilution
lowered the mean hematocrit 9 and 13%, respectively, from a mean of 5 1.5% (Table). Following
stage 1 and stage 2 hemodilution, an increase in gray
matter flow of 19 and 23%, respectively, was observed (Fig 2). Arterial oxygen content (Tao,) decreased, as could be expected, but oxygen delivery
capacity (CBF x Tao,) to the brain remained constant due to the CBF increase (Table). The oxyhemoglobin dissociation curve was unchanged as
P,, remained constant. The cerebral metabolic rate
mean homatocrit
F i g I . lndiuidual cerebral blood f l o w responses in six patients
with polycythemia following a rapid, normouolemic two-stage
for oxygen (CMRO,) was slightly increased during
the first stage of hemodilution but returned to control level during the second stage (Table). Arterial
carbon dioxide tension (Paco,) showed a minor and
insignificant decrease, less than 2 mm Hg, which was
probably dye to the slightly acidotic p H of the albumin solution, a factor which was corrected for in the
later studies. The jugular venous partial pressure of
oxygen (Pvo,) fell slightly, 1.6 mm H g on average,
following stage 1 and 2 hernodilution (see Fig 2); it
remained constant, however, in those patients in
whom Paco, was unchanged. The patients were all
normotensive prior to and following the isovolemic
hemodilution; thus, increases in blood flow were
paralleled by corresponding decreases in cerebrovascular resistance.
Response to Hemodilution in Six Patients with High Hematocrit
Hematocrit (%)
Cerebral blood flow
(m11100 gmimin)
Venous partial pressure
of oxygen (mm Hg)
Arteriovenous oxygen
content difference (ccidl)
Cerebral metabolic rate
for oxygen (cci100 gmimin)
Arterial oxygen content (ccidl)
Oxygen delivery capacity
(cci100 gmimin)
Partial pressure
of carbon dioxide (mm Hg)
Mean (Range)
Stage 1
(mean t ASD)”
2 1.4
Stage 2
(mean ? ASD)”
* 1.7
p < 0.01
p < 0.01
51.5 (48-54)
44.2 (37-60)
52.7 2 3.3
35.2 (30-30)
33.6 -+ 2.3
7.12 (5.2-8.2)
6.70 f 0.7
5.92 t 1.1
p < 0.01
3.0 (2.4-3.9)
3.5 t 0.5
3.2 -+ 0.6
p < 0.05
21.7 (20-23)
9.6 (7.7-11.9)
19.5 ? 0.8
10.2 t 0.7
18.5 t 1.5
10.0 t 1.2
p < 0.01
39.1 (36-44)
t 2.0
37.7 t 2.1
aASD = standard deviation of changes from baseline.
Henriksen et al: CBF following Venisection
1E L t
Thomas et a l . (1977)
mean hernatocrit
Fig 2. Influence of rapid, normovolemic two-stage hemodilution
on cerebral hloodjow and blood gases. VertiLa1line.f indicate
standard deviation ofthe differences compared t o control value.
Blood viscosity and oxygen release capacity both
seem to be involved in the regulation of CBF. The
viscosity of blood is mainly attributable to the presence of erythrocytes (viscosity increases with high
hematocrit and decreases with low hematocrit), although other factors, such as plasma viscosity, also
influence the viscosity of blood. T h e oxygen release
capacity of blood is determined by its oxygen binding
capacity as well as by the ability of oxyhemoglobin to
release oxygen. The oxygen release capacity can thus
be defined as the amount of oxygen that can be released from fully oxygenated (saturated) blood when
the PO, is dropped to a given value, e.g., to 35 mm
H g (a value close to the Pvoz when a tissue is adequately supplied with oxygen). In polycythemia, viscosity and oxygen release capacity are increased and
CBF is reduced [ l , 3, 4 , 11;12], whereas in anemia,
viscosity and oxygen release capacity are decreased
and CBF is increased [ S , 7-10]. Discrepancies have
been observed, however. Wade and co-workers 1131
reported increased CBF despite high blood viscosity
in a group of patients with elevated hematocrit and
with a high-oxygen-affinity hemoglobin variant (i.e., a
relatively low oxygen release capacity compared to
the hematocrit). In patients with paraproteinemia
(resulting in increased blood viscosity), Humphrey
et a1 [2] found a CBF that was comparable to normal values despite the presence of anemia. CBF was
significantly lower than in patients who had anemia
without increased viscosity. Paulson and associates
"71 attempted to dissociate the effect of oxygen re-
456 Annals of Neurology Vol 9 No 5 May 1981
F i g 3. Cerebral blood f l o w changes obtained in various jtudies
following reduction of hematocrit using venzjectzon alone or
follou'ed bji volume replacement with an albumin or plasma
lease capacity and blood viscosity by hemodilution
and by carbon monoxide inhalation, and found a
dominant role of both oxygen release capacity and
viscosity. The two factors thus seem to be involved in
regulation of CBF, but independently.
Thomas and colleagues [ l l , 121, in two separate
studies, observed an increase in CBF of 50 and 7396,
respectively, following moderate hemodilution in
patients with high-normal hematocrit (Fig 3). T h e
CBF increase was much more pronounced than the
decrease in hematocrit, suggesting a state of relative
generalized cerebral hypoxia. In contrast, the main
finding of the present study was that the increase in
CBF following hemodilution corresponded to the
decrease in hematocrit (Fig 3) since oxygen delivery
capacity (CBF x Tao,) remained constant. Furthermore, one would have expected the flow increase to
be accompanied by an increase in jugular Pvo2 if there
was a state of relative cerebral hypoxia when hematocrit was high. However, we observed that the
jugular Pvo2 remained constant. The discrepancy
between the present study and tbqr of Thomas et a1
[ I 1, 121 is difficult to explain, altljnugh several factors might be involved. They did not measure jugular
Pvo2. We measured the immediate effect of
hemodilution, whereas days to weeks elapsed in their
studies, and other unknown factors might have contributed to the observed CBF increase. The method
for CBF measurement was also different: we used the
intraarterial '":'Xe injection method, whereas they
used the atraumatic intravenous modification of the
method, which introduces some errors such as contamination by extracranial blood flow, radiation from
xenon trapped in the frontal sinuses, and cross-talk
(measuring activity from the contralateral hemisphere as both are supplied by '""Xe).
Both arterial and jugular venous blood samples
were collected in the present study, which allowed us
to take metabolic events into consideration, although
some precautions are needed in interpreting the results. CMRO, calculated from the initial CBF (mainly
cortical gray matter) and arteriovenous oxygen content differences, (A-V)o, (whole brain), does not
reflect whole-brain metabolism but approximates
CMRO, in gray matter. The same criticism can be
applied to the calculated oxygen delivery capacity
(CBF x Tao.,). C M R 0 2 increased significantly, by
1796, during stage 1 hemodilution, but returned to
the baseline level following stage 2 (see the Table).
This could be a statistical coincidence, although it
seems unlikely since all patients reacted the same way
following stage 1 hemodilution. If we consider the
equation (A-V)o, = CMROJCBF and assume that
CMRO, of whole brain remained constant, it can be
calculated that total CBF increased by 6 and 2096,
respectively, following the two stages of hernodilution. However, the observed increases in initial CBF
were 19 and 23%, respectively. The discrepancy
between the C M R 0 2 and CBF changes following
stage 1 hemodilution might reflect unchanged or diminished flow in parts of the brain other than those
reflected in initial CBF values (heterogeneity); alternatively, it might be due to a real increase in CMRO,
caused by some unknown mechanism. However, no
matter which events are involved in these metabolic
findings, they d o not support a marked CBF increase
with improved cerebral oxygenation.
In conclusion, the present investigation confirms
that CBF is reduced in patients with high-normal
hematocrit. We have not been able to reproduce the
findings of Thomas and colleagues [ 11, 121 that CBF
rises markedly following hemodilution: on the con-
trary, we observed a flow increase that matched the
decrease in hematocrit and oxygen release capacity,
as evidenced by an unchanged jugular Pvo2.
Supported by the Danish Medical Research Council and the
Danish Heart Association.
1. Gottstein U: Cerebral blood flow, C M R 0 2 and CMR of glucose in patients with hypo- and hyperchromic anemia and
polycythemia. The effect of hemodilution on CBF, CMR and
hematocrit. In Meyer JS, Lechner H , Reivich R (eds): Cerebral Vascular Diseases 2. Amsterdam and Oxford, Excerpta
Medica, 1978
2. Humphrey PRD, du Boulay GH, Marshall J, et al: Viscosity,
cerebral blood flow and haematocrit in patients with paraproteinaemia. Acta Neurol Scand 61:201-209, 1980
3. Humphrey PRD, Marshall J, Ross Russell RW, et al: Cerebral
blood flow and viscosity in relative polycythemia. Lancer
2:873-877, 1979
4. Merrill EW: Rheology of blood. Physiol Rev 49:863-888,
5. Michenfelder J, Theye R: The effect of profound hypocapnia
and dilutional anemia on canine cerebral metabolism and
blood flow. Anesthesiology 31:440-457, 1069
6.Olesen J, Paulson OB, Lassen NA: Regional cerebral blood
flow in man determined by the initial slope of the clearance of
intraarterially injected xenon-133. Stroke 2:5 19-540, 1971
7. Paulson O B , Parving H , Olesen J, e t al: Influence of carbon
monoxide and of hemodilution on cerebral blood flow ancl
blood gases in man. J Appl Physiol 35:111-116, 1073
8. Rosberg B, Wolff K: Regional blood flow in normovolaemic
and hypovolaemic haemodilution. Br J Anaesth 5 1:423-430,
9. Rosenblum WI: Effects of reduced hematocrit on erythrocyte
velocity and fluorescein transit time in the cerebral microcirculation of the mouse. Circ Res 24:96-103, 1971
10. Shimoji K, Maruyama M, Higuchi A: Effect of hemodilution
on autoregulation in cerebral blood flow. Acta Neurol Scand
[Suppl] 72:620-621, 1979
11. Thomas DJ, Marshall J, Ross Russell RW, e t al: Cerebral
blood-flow in polycythemia. Lancer 2:161-163, 1977
12. Thomas DJ, Marshall J, Ross Russell RW, et al: Effect of
hematocrit o n cerebral blood-flow in man. Lancer 2 9 4 1-943,
13. Wade JPH, du Boulay G H , Marshall J, et al: Cerebral blood
flow, haematocrit and viscosity in subjects with a high oxygen
affinity haemoglobin variant. Acta Neurol Scand 61:2 10-21 5,
Henriksen et al: CBF following Venisection
Без категории
Размер файла
372 Кб
flow, patients, hematocrit, following, high, hemodilution, cerebral, normavolemic, blood
Пожаловаться на содержимое документа