A Study of Hepatic Hemodynamics
in the Dog
By Charles P. Shoemaker, Jr., M.S.
Downloaded from http://circres.ahajournals.org/ by guest on June 17, 2017
• The purpose of this work was to study in
detail some of the factors related to the rise
in portal vein pressure which accompanies
sympathetic nerve stimulation. These hemodynamic factors were studied by observing
the changes in hepatic artery and portal vein
vascular resistances noted during controlled
perfusion of the liver in vivo. Initially, the
pressure-flow characteristics of the hepatic
artery and portal vein vasculature and the interrelationship between the two vessels were
studied in the nonstimulated state; then these
same parameters were restudied during and
following stimulation of sympathetic innervarion.
The rise in portal vein pressure of dogs as
a result of splanchnic nerve stimulation has
been known since 1894.1 In the resting state
portal vein pressure depends upon three factors — portal flow, the vascular resistance to
portal flow through the liver, and pressure in
the vena cava. Portal flow in turn depends
upon the vasomotor tone in the mesenteric
and splenic beds, portal venous pressure, and
systemic blood pressure. Previous results
shown in table 1 and figure 1 indicate how
these and other hepatic parameters respond
to sympathetic stimulation. It was noted by
Green et al.2 that vena cava pressure was not
altered markedly by splanchnic nerve stimulation, yet portal pressure rose. This suggests
This work was done under joint supervision of the
Departments of Physiology and Experimental Surgery, Albany Medical College, Albany, New York.
Mr. Shoemaker held a medical student research
Fellowship from the Heart Association of Albany
County, Inc.
Submitted in partial fulfillment of the requirements for the M.S. degree.
Supported by Grant H-4382 from the U. S. Public
Health Service.
Received for publication February 10, 1984.
216
that the rise in portal pressure must be due
to a rise of either portal flow or portal vein
vascular resistance. Green et al.2 has shown
also that portal flow decreases during the initial phase of nerve stimulation. Consequently,
since vena cava pressure does not change and
portal flow decreases, the initial rise in portal
vein pressure is probably due to a rise in
portal vein vascular resistance.
Two series of in vivo perfusions of canine
liver were undertaken. In Series I only the
hepatic artery was perfused and the portal
FiGURE 1
Diagram summarizing changes of licer volume, mesenteric vascular resistance, portal blood flow, mean
arterial blood pressure (MABP), portal vein pressure
(PVP), and inferior vena cava pressure (IVCP)
noted with splanchnic nerve stimulation. Heavy solid
line indicates period of stimulation. Data derived
from published studies shown in table 1.
Circulation Research, Vol. XV, September 1964
217
HEPATIC HEMODYNAMICS
TABLE 1
Hepatic Heinodynamics: Summary of Changes Noted During Sympathetic Stimulation
by Previous Investigators
Change
Parameter
Investigator
Downloaded from http://circres.ahajournals.org/ by guest on June 17, 2017
Hepatic artery flow
increases
Green et al.2
Systemic blood pressure
increases
Green et al.2
Mesenteric vascular
resistance
constriction
followed by
vasodilatation
Deal and Green 3
Portal flow
decreases
followed by
an increase
Green et al.2
Portal pressure
increases
Green et al. 2
Diameter of
intrahepatic
sinusoids
decreases
Seneviratne,4
Wakim 5
Liver volume
decreases
Griffith and Emery 0
decreases
(without spleen)
Green et al.2
increases
Macleod and Pearce 7
no change
Green et al.2
Total hepatic blood flow
1) inflow
2) outflow
Vena cava pressure
system was left intact. In Series II, both the
hepatic artery and portal vein were perfused.
In both sets of experiments the effects of altered hepatic artery flow on hepatic artery
perfusion pressure and portal vein pressure
were studied. When both vessels were perfused, the effects of altered portal vein flow on
portal pressure and hepatic artery hemodynamics were also noted. In this latter
preparation flows through the hepatic artery
and portal vein could be held constant, and
the effects of drugs or of nerve stimulation on
the vascular resistance were readily determined by observing the changes of perfusion
pressures.
Methods
A schematic diagram of the combined hepatic
artery and portal vein perfusion (Series II) is
shown in figure 2. The experimental setup used
in Series I was similar except that the portal
circulation was left intact. Since many of the
techniques in the two series were similar, several
of the common features will be described initially,
then the technique of perfusing the hepatic artery
Circulation Research, Vol. XV, September 1964
(Series I) will be presented, and finally the
additional steps required for combined perfusion
(Series II) will be described.
Special efforts were made to avoid the condition known as "outflow block,"8 which in dogs is
believed to be due to spasm of the hepatic vein.9
This can be caused by histamine, undue manipulation of the viscera, hepatic anoxia, transfusion
reaction, or shock.8'10 When "outflow block" is
present, the livers become congested, swollen,
and cyanotic. One precaution taken to prevent
"outflow block" was to prepare all cannulae and
recording devices before the abdominal cavity
was entered in order to reduce the time die
viscera were exposed. Second, in all surgical dissections any unnecessary manipulation of the
liver was avoided. Third, in order to prevent
hepatic anoxia, the cannulations of the hepatic
artery were done in a definite order so that
hepatic artery flow was never interrupted. As a
final precaution isotonic saline was the only fluid
administered. Blood transfusions were not used
because of the known hepatic response to transfusions.0
Dogs weighing 13 to 27 kg with livers weighing 340 to 1,000 g were anesthetized with
sodium pentobarbital (30 mg/kg). The femoral
vessels were isolated bilaterally. One femoral
218
SHOEMAKER
HEPATIC ART
^
PRESSURE
PORTAL VEIN
J\. RECORDING
DEVICE
MESENTERIC
V^
@@ @ ©
—i
i_
HEPATIC
ARTERY
SHUNT
Downloaded from http://circres.ahajournals.org/ by guest on June 17, 2017
CHR
I
PORTAL VEIN
PUMP
STOPCOCK"^
FIGURE 2
Schematic diagram of perfusion circuits. PCS = portacaval shunt, CHR = constant height
reservoir. Oilier symbols defined in text.
artery was cannulated in order to measure systemic blood pressure, one vein was cannulated
for administration of fluids and for measuring
pressure in the vena cava, and the other artery
and vein were prepared for future cannulation.
Vena cava and portal vein pressures were measured with either a spinal manometer or a strain
gauge, and arterial pressures were measured with
either mercury manometers or strain gauges. In
all cases zero pressure was located at the level of
the groin. The perfusion pressures reported were
the mean of the recorded pressures, which were
sinusoidal in shape due to the nature of the
pump.
SERIES I, INTACT PORTAL VEIN
The abdomen was opened through a midline
incision. A catheter was inserted into a small
mesenteric vein (F, fig. 2) and was threaded into
a larger mesenteric vein in order to measure
portal venous pressure. The central end of the
hepatic artery (B) was dissected free for future
cannulation, and then the distal extension of the
common hepatic artery (G) was cannulated. This
cannula in the distal end of the hepatic artery
(G) was inserted centrally towards the aorta so
that it could serve as a temporary hepatic artery
shunt while the central end of the hepatic artery
was cannulated to complete the perfusion circuit.
After perfusion was established, this cannula
(G) was disconnected from the perfusion circuit
and then used as a catheter to measure pressure
in the hepatic artery. This distal cannula was
inserted only a few millimeters to prevent occluding any of the individual hepatic arteries. Before
this cannula was fixed in position by ligatures, it
was adjusted so that blood pressures measured
simultaneously in this catheter and the femoral
artery were nearly equal. Since these two pressures were nearly equal, it indicated that between the aorta and this distal catheter (G)
there was no significant vascular resistance due
to twisting or kinking of the hepatic artery or to
malposition of the catheter and that the pressure
recorded in this distal catheter equaled hepatic
artery pressure. Because narrow cannulae introduce artificial vascular resistances, only those
perfusion pressures recorded distal to the perCircmltiion Rentrcb, Vol. XV, September 1964
HEPATIC
HEMODYNAMICS
Downloaded from http://circres.ahajournals.org/ by guest on June 17, 2017
fusion cannulae were used in calculating the vascular resistances. After the dissection was completed, mass ligatures were placed around all
other structures in the lesser omentum except for
the portal vein, hepatic artery, and hepatic nerves
in an effort to prevent leakage through collaterals.
The dog was then heparinized using 1000 USP
units/6 lb.
The hepatic artery circuit was primed with
systemic arterial blood after cannulation of the
femoral artery (A). A temporary circuit was
made by connecting the outflow from the perfusion pump to the femoral vein. Circulation
through this circuit was continued for ten minutes
to allow any vasoactive substances to be washed
out of the tubing. This perfusion circuit was then
disconnected from the femoral vein and connected to the cannula in the distal end of the
hepatic artery (G) to form a temporary hepatic
artery shunt. Perfusion of the hepatic artery with
femoral artery blood was then started through
this shunt at a rate of 60 cc/min. The exposed
proximal end of the hepatic artery (B) was next
cannulated and connected to the perfusion circuit by a Y-tube so that the hepatic artery was
being perfused at this time from both ends. The
distal cannula (G), which was being used as a
temporary shunt, was then disconnected from the
Y-tube in the arterial perfusion circuit and was
connected to a mercury manometer in order to
measure perfusion pressure within the hepatic
artery. In addition, perfusion pressure within the
perfusion cannula was recorded by means of a
catheter connected to the Y-tube in the perfusion
circuit. Hepatic artery flow was determined and
regulated by the setting of the Sigmamotor pump,
which was calibrated at the end of the perfusion. This pump would maintain calibrated flow
only if the blood pressure in the cannula entering
the pump was held constant by a constant height
reservoir shown in figure 2.
Before terminating an experiment, the hepatic
nerves were transected and the distal ends connected to stimulating electrodes. The nerves were
stimulated with pulses having a duration of 70
msec and a frequency of 3 to 10 pulses/sec. The
voltage was varied from 5 to 30 volts. In an
acceptable experiment, stimulation produced typically a threefold increase in hepatic artery resistance indicating that the nerves were intact
between the electrodes and the vascular bed and
that the vascular bed was still capable of marked
vasoconstriction.
SERIES I I , COMBINED PORTAL VEIN AND
HEPATIC ARTERY PERFUSION
In this series of experiments the hepatic artery
and portal vein were perfused by two extracorporeal circuits as diagrammed in figure 2. The
Circulation Rtitsrcb,
Vol. XV, Stpitmher 1964
219
hepatic artery was perfused as in Series I. The
second circuit drained total mesenteric blood via
the splenic vein (C) into a 400 cc beaker
mounted on a strain gauge, after the portal vein
had been ligated distal to the junction of the
mesenteric and splenic veins (D). This mesenteric blood was pumped out of the beaker by the
larger Sigmamotor pump into the hepatic end
of the ligated portal vein (E). Flows through the
hepatic artery and portal vein were altered and
regulated by setting the speeds of the Sigmamotor pumps, which were calibrated at the end
of each perfusion. Mesenteric blood flow, which
was essentially determined by systemic arterial
pressure and mesenteric vascular resistance, was
calculated by adding gains or losses of blood
weight in the beaker on the strain gauge to the
portal flow through the large Sigmamotor pump.
The strain gauge was arranged to measure
changes in the weight of the beaker, and these
changes were recorded with a Sanborn recorder.
In three of the dogs, after the femoral vessels
had been cannulated, the right splanchnic nerve
was exposed through a right thoracotomy at the
tenth interspace while respiration was maintained
with positive pressure oxygen. The nerve was
transected and the distal portion was fastened to
silver stimulating electrodes. Intrapleural air was
removed through a chest catheter using an underwater seal, the wound was closed, and artificial
respiration was discontinued. The abdomen was
then opened as before, a mesenteric vein was
catheterized to measure mesenteric venous pressure, the proximal and distal ends of the hepatic
artery were again dissected, and a cannula (G)
was inserted centrally into the distal end of the
hepatic artery. A large branch of the splenic vein
(C) was dissected free for later cannulation. The
splenic arteries were ligated near the spleen, the
blood allowed to flow out of the spleen, and the
spleen removed. A one-inch length of the portal
vein (E) proximal to its bifurcation was also dissected free. The pancreaticoduodenal vein was
ligated and transected. Again mass ligatures were
placed around nonessential structures in the lesser
omentum and the dog heparinized.
The hepatic artery circuit was primed with
blood, the tubing washed out by temporarily circulating the blood back into the femoral vein,
and then the hepatic artery was perfused through
both the proximal and distal ends of the artery
as described above in Series I. The splenic vein
(C) was cannulated with a large bore cannula.
This cannula led to the beaker on the strain gauge
for recording flow. The portal vein (D) was then
ligated distal to the junction of the splenic vein
with the mesenteric vein. The level of the beaker
was adjusted so that the mesenteric venous pressure was near 20 cm of saline. Blood from this
220
SHOEMAKER
Downloaded from http://circres.ahajournals.org/ by guest on June 17, 2017
beaker was then pumped by the Sigmamotor pump
into a femoral vein (H), creating a temporary
portacaval shunt. A cannula, connected by a
Y-tube (I) to the shunt between the pump and
the femoral vein, was then inserted and secured
in the hepatic end of the ligated portal vein (E).
Mesenteric flow, which was being pumped by the
large Sigmamotor pump into the femoral vein, was
then diverted to the liver by clamping the
femoral vein cannula (H). The perfusion pressure
of this portal flow was measured by a catheter,
extending about 3 mm beyond the opening of
the cannula. After the portal perfusion was
started, the temporary hepatic artery shunt via
the cannula in the distal end of the hepatic artery
(G) was disconnected and used as a catheter
for measuring hepatic artery perfusion pressure
as in Series I. By regulating the large Sigmamotor
pump the blood level in the beaker did not vary
more than 5 cm, and thus the pressure head
supplying this pump was held nearly constant so
that it would maintain a constant output for a
given setting.
After splanchnic nerve stimulation, the hepatic
nerves were transected and stimulated distally,
using the same stimulus as before. As mentioned
above, evidence of satisfactory stimulus, and intact nerves, was the marked rise in hepatic artery
resistance with stimulation (fig. 8).
Having determined the various flows and the
corresponding pressure gradients, the following
vascular resistances could be calculated:
Mesenteric resistance =
Systemic arterial press. — Mes. vein press.
Mesenteric venous flow
Portal venous resistance =
PV perfusion press. — Vena cava press.
Portal vein flow
Hepatic arterial resistance =
HA perfusion press. — Vena cava press.
Hepatic artery flow
The resistances were measured 2 in peripheral
resistance units (PRU = mm Hg/cc/min).
At the end of all experiments, the extrahepatic
blood vessels were examined to ascertain that no
aberrant vessels were present and that all catheters and cannulae were in their correct positions.
After the blood had been allowed to drain from
the large hepatic vessels, the livers were weighed
with the gall bladders attached.
Results
Six livers were satisfactorily perfused via
the hepatic artery in Series I and ten livers
via both hepatic vessels in Series II. At the
end of the perfusions the livers were noted
to be soft and red in all cases in Series I and
in all but one case in Series II. This one exception, in which the liver became blue and
congested, has been retained and reported because the results were qualitatively similar to
those obtained from the other perfusions.
The durations of the perfusions varied, but in
no case was a perfusion terminated due to
increasing vascular resistance. On the contrary, the vascular resistance of both the
hepatic artery and portal vein decreased as
the perfusion continued. The systemic pressure and other hemodynamic parameters recorded are listed in table 2.
EFFECT OF ALTERED HEPATIC ARTERY FLOW: SERIES I
A typical result obtained when hepatic artery flow was altered, with portal vein left
intact, is shown in figure 3. Flow through the
hepatic artery was changed in a step-like
fashion over a period of three to five minutes. As flow was altered, the following were
recorded: portal vein pressure, mean arterial
blood pressure, hepatic artery perfusion pressue measured by means of the catheter in the
distal hepatic artery (G, fig. 2), and "perfusion cannula pressure" measured by means of
a catheter connected to the Y-tube in the
perfusion cannula between the pump and the
artery. It was noted in all dogs that variations
in hepatic artery flow as much as from 50 to
200 cc/min increased the portal pressure by
only 2 cm of saline. As shown in figure 3
the rise in "perfusion cannula pressure" was
much greater than the rise in the hepatic artery pressure recorded in the distal hepatic
artery. This marked difference between the
two arterial pressures was noted in all cases,
showing that along the arterial outflow from
the perfusion pump there was a conspicuous
pressure gradient between the Y-tube in the
perfusion cannula and the end of the pressure
catheter, which had been placed retrograde
in the distal end of the hepatic artery. Since
this catheter in the distal hepatic artery had
been adjusted so that it recorded systemic
pressure accurately prior to the perfusion, it
was therefore considered to give a true measCircidMiio* Reienrcb, Vol. XV, Septtmber
1964
221
HEPATIC HEMODYNAMICS
TABLE 2
Averaged Results for Each Dog
During perfuiion
Dog
no.
Do*
weight
Liver
weight
Before perfusion
MABP*
Portal vein
pressure
MABP*
Portal vein
Pressure
Flow
cm HiO cc/min
Mesenteric
vein
pressure
Vena
cava
pressure
cc/min
cm HaO
cm HiO
210
240
320
97
320
246
—
—
—
—
—
—
—
—
50
130
80
60
50
140
75
96
80
90
15
18
Hepatic artery
Perfusion
pressure
Flow
mm Hg
mm Hg
cm HaO
mm Hg
—
—
—
—
—
—
110
130
150
140
80
110
16
11
16
15
22
13
—
—
100
100
100
100
100
100
100
90
85
50
100
90
90
90
80
110
23
12
25
20
16
12
15
15
10
13
120
150
170
120
152
180
170
236
265
187
65
90
60
70
50
60
40
45
70
70
-
Series I
1
2
3
4
5
6
13.5
20
21
13.5
16.0
13.5
600
550
800
340
570
425
—
—
—
—
—
—
20
20
16
20
20
25
27
27
20
27
550
520
120
100
—
—
150
180
130
120
125
120
—
Series II
Downloaded from http://circres.ahajournals.org/ by guest on June 17, 2017
7
8
9
10
11
12
13
14
15
16
600
690
760
1000
760
920
18
15
17
20
20
11
15
20
20
20
25
20
20
15
11
10
8
12
8
6
* MABP = mean arterial blood pressure.
ure of the pressure within the hepatic artery
during the perfusion. Consequently, the pressure gradient between the Y-tube and the distal hepatic artery was ascribed to an artificial
resistance between the perfusion cannula and
the hepatic artery due to the small diameter
of the perfusion cannula and failure to obtain
perfect alignment of the perfusing cannula in
the artery. Thus, unless specified otherwise,
hepatic artery pressure means that pressure
which was measured in the distal end of the
hepatic artery.
Average hepatic artery pressure-flow curves,
as shown in figure 4, were then constructed,
from results such as those in figure 3, by
plotting the perfusion pressures measured
with the catheter in the distal hepatic artery
against the blood flows. It was noted that if
additional pentobarbital had been injected,
there was marked vasodilatation for at least
15 minutes following the injection. All the
pressure-flow curves for a given dog, except
for those curves obtained within 15 minutes
after the administration of pentobarbital, were
Circulation Rtietrcb, Vol. XV, Sipumbtr
1964
then averaged to yield the average curves
shown in figure 4. It was found that these
average curves were linear between 40 and
80 mm Hg. The individual curves were extrapolated to 100 mm Hg in those cases in
which the perfusion pressure was less than
100 mm Hg at the maximal flow of the perfusion pump. The average hepatic artery
flows at 100 mm Hg for each dog and the
corresponding body and liver weights are
presented in table 2. During two of the perfusions hepatic artery flow was stopped momentarily, and hepatic artery pressure fell
towards, but remained above, portal pressure.
Average portal pressure was about 15 mm
Hg, whereas the hepatic artery pressure at
zero flow was about 20 mm Hg.
EFFECT OF ALTERED HEPATIC ARTERY FLOW: SERIES II
In those experiments in which both vessels
were perfused, the hepatic artery flow was increased and decreased gradually while the
portal flow was held constant in nine dogs.
The changes noted in all dogs were similar to
222
SHOEMAKER
H A
FL OVSJ.
2 50
a>
1200 _
PCRFUSION
1
CANNUt A
PRESS
150
-—
i1
flf
r*
\L"
\
-
"100
—
S^
\
at
"• 50
A
Downloaded from http://circres.ahajournals.org/ by guest on June 17, 2017
r
O
s 10
PRESS
50
30
20
P V P
10
o
o
5
1
•
4
•
m i n
20
40
60
BLOOD
FIGURE 3
Results of typical experiment, noted with dog 6 of
Series I, showing effects of altered hepatic artery flow
on portal vein pressure (PVP), hepatic artery pressure (HA PRESS, recorded at point C, fig. 2), and
hepatic artery "perfusion cannula pressure" (recorded
by means of Y-tube inserted in perfusion cannula
between pump and artery).
the results of clog 16 shown in figure 5. A
twofold increase in hepatic artery flow resulted in less than a twofold increase in
hepatic artery profusion pressure (distal
hepatic artery). Portal pressure rose only
slightly. During these increases of hepatic
artery flow, portal flow was held constant,
hence portal pressure was directly proportional to portal vein resistance. Hepatic artery
flow was then altered after the portal blood
had been shunted by closing the portal vein
limb of the Y-tube (I, fig. 2). The pressure
recorded with the pressure cannula still in the
portal vein stump after shunting is shown in
figure 5 as "stump" pressure. This "stump"
pressure paralleled the previous portal pressure changes even though portal flow had
been diverted around the liver. The hepatic
artery pressure-flow curve constructed from
these data was not altered markedly when
FIGURE 4
Individual average hepatic artery pressure-flow curves
of six dogs in Series I.
hepatic portal flow was reduced to zero by
shunting (fig. 6). A slight rise in portal pressure and "stump" pressure accompanied the
increased hepatic artery flow.
EFFECT OF ALTERING PORTAL FLOW WITH
CONSTANT HEPATIC ARTERIAL FLOW
In several dogs of Series I, in which only
the hepatic artery was perfused, the portal
vein was ligated acutely for 30 seconds by
pulling up on a ligature which had been
placed around the portal vein. With a fixed
hepatic artery flow, acute ligation of the portal vein caused a 20% decrease in hepatic
artery perfusion pressure, indicating that a
sudden fall in portal vein flow reduces hepatic
arterial resistance. After 30 seconds, the ligature was released and the hepatic artery perfusion pressure rebounded to slightly above
control levels, but returned to normal in one
minute. Systemic pressure was not appreciably altered.
In Series II, in which both vessels were
perfused, portal flow to the liver was altered
Circulation Reittrcb, Vol. XV, Stpttmbtr 1964
223
HEPATIC HEMODYNAMICS
250
60
Downloaded from http://circres.ahajournals.org/ by guest on June 17, 2017
PORTAL
o
o
~TO
i
CO
SHUNT
LIVER
P VP
10.0
O
PORTACAVAL
FLOW
STUMP"
V P
1
!
mi n
FIGURE 5
Results of typical experiment, noted with dog 16 of Series II, showing effects of altered
hepatic artery flow on hepatic artery (HA) perfusion pressure, vena cava pressure, portal vein
pressure before portacaval shunt, and "stump" vein pressure ("stump" VP) during portacaval
shunt. Portal flow to liver = 185 cclmin. Symbols otherwise as in Figure 1.
20
40
60
HEPATIC
80
100
ARTERY
120
140
160
BLOOD
FLOW
180
200
cc/min
220
FIGURE 6
Changes in hepatic artery perfusion pressure, portal pressure, "stump" pressure, and vena
cava pressure noted in figure 5 as a function of hepatic artery flow. Symbols as in figures
1 and 5. o—o Hepatic artery pressure before portacaval shunt. • — • Hepatic artery perfusion
pressure after portacaval shunt.
CircuUtion Rtittrcb, Vol. XV. S.pitmbtr 1964
224
Downloaded from http://circres.ahajournals.org/ by guest on June 17, 2017
either by shunting or by altering rates of
blood flow. In some cases this alteration in
portal flow had no effect on hepatic artery
resistance, as shown in figure 6 where the
hepatic artery pressure-flow curve was essentially the same before and after shunting. An
exception to this generalization is shown in
figure 7 from dog 9, in which portal flow was
varied over a very wide range and hepatic
artery resistance varied directly with the rate
of portal flow to the liver. The splanchnic
and hepatic nerves were intact and it is conceivable that some systemic response may
have accounted for this change in hepatic
resistance, since systemic arterial pressure
varied in parallel with hepatic arterial perfusion pressure. Alternatively, these changes
in hepatic resistance may be related to intrahepatic congestion secondary to the changes
in portal flow. In one instance (dog 16) when
portal flow was diverted to the liver for the
first time, there was a transient increase in
hepatic artery resistance which soon returned
to the level existing before portal flow was
passing through the liver. Otherwise, hepatic
artery resistance showed very little response
to alterations in portal flow. Usually as the
perfusion progressed, the apparent hepatic artery resistance decreased and became less
responsive to changes in portal flow.
EFFECTS OF SPLANCHNIC NERVE STIMULATION
Studied in three dogs of Series II, splanchnic nerve stimulation caused an initial decrease in mesenteric blood flow, followed by
an increase, and a rise in hepatic artery and
portal vein perfusion pressures (fig. 8). During portacaval shunt, stimulation raised
"stump" pressure. Changes in the portal vein,
mesenteric, and hepatic artery vascular resistances indicated vasoconstriction in all
three beds (fig. 9). The initial vasoconstriction could be prolonged with longer periods
of stimulation. In the earlier part of the perfusions splanchnic stimulation was followed
by secondary vasodilatation in the mesenteric
as well as in the hepatic artery and portal
vein beds. However, toward the end of the
perfusion the secondary dilatation of the
SHOEMAKER
430
j
3
FIGURE 7
Results of typical experiment, noted with dog 9 of
Series II, showing effects of altered portal vein flow
on hepatic artery perfusion pressure, portal pressure,
and vena ctwa pressure. Since hepatic artery flow was
fixed at 80 cc/min, hepatic artery perfusion pressure
paralleled hepatic artery vascular resistance. Symbols as in figures 1 and 5.
hepatic artery and portal beds was absent.
In those cases showing secondary vasodilatation of the hepatic artery and portal vein following stimulation, the control (nonstimulated) resistances were usually higher than
the control (nonstimulated) resistances later
in the perfusion.
STIMULATION OF THE HEPATIC NERVES
The distal ends of the transected hepatic
nerves were stimulated in four dogs of Series
I. This increased hepatic artery perfusion
pressure threefold and elevated portal vein
pressure by 3 or 4 cm of saline. Stimulation
of the peripheral end of the divided hepatic
nerves in Series II (four dogs) caused
changes which generally were limited to the
liver (fig. 10). These changes were similar to
splanchnic stimulation, in that both hepatic
Circulmtion Rtsetrcb, Vol. XV, Stpttmbtr 1964
225
HEPATIC HEMODYNAMICS
Downloaded from http://circres.ahajournals.org/ by guest on June 17, 2017
FIGURE 8
Results of typical experiment, noted with dog 15 of Series II, showing effects of right
splanchnic nerve stimulation on mesenteric blood flow, systemtc blood pressure, vena cava
pressure, portal pressure before portacaval shunt, and "stump" vein pressure during portacaval
shunt. Hepatic artery flow was fixed at 80 cc/min. Symbols as in figures 1 and 5.
c
i 16
X
HEPATIC
> ARTERY
y«v
I
i
6 12
E
UJ
V
10
u
Z
^o.
m 0.6
EFFECT OF EPINEPHRINE
>MESENTER IC
\
c
<S
y
0.03
PORTAL
VEIN
002
0.01
1
^
1
1
FIGURE 9
Effects of right splanchnic nerve stimulation on hepatic artery vascular resistance, mesenteric vascular
resistance, and portal vein vascular resistance. Resistances calculated from data in figure 8.
Circulation Rtftrcb, Vol. XV, Slpttmbtr 1964
artery and portal vein resistances increased,
"stump" pressure rose after the portal blood
had been shunted to the femoral vein, and
there was a secondary dilatation when the
prestimulation resistances were high. These
results are shown in figures 8 and 10.
At the conclusion of several experiments
epinephrine was injected. In two dogs in
which only the hepatic artery was perfused
0.25 mg injected into the portal vein increased portal vein pressure and increased
hepatic artery resistance. In one of these two
dogs the injection of 0.25 mg into the hepatic
artery increased hepatic artery resistance
markedly and increased portal vein pressure
slightly. Epinephrine was also injected into
two dogs in which the liver was perfused by
both vessels. In dog 13, 0.1 mg was injected
into the hepatic artery, and in dog 14, 0.25
mg was injected into the portal blood going
to the liver. With hepatic artery and portal
flows fixed, epinephrine elevated hepatic ar-
SHOEMAKER
226
200
160
at
1
160
H A
PERFUSION
PRESS
E
E 140
... ' 2 0
o.
O
O
80
Downloaded from http://circres.ahajournals.org/ by guest on June 17, 2017
"PORTAL
FLOW TO LIVER
P VP
FIGURE 10
Results of typical experiment, noted with dog 16 of
Series II, showing effect of hepatic nerve stimulation
on hepatic artery perfusion pressure, portal pressure
before shunt, and "stump" vein pressure during
portacaval shunt. Hepatic artery flow = 152 cc/min,
portal vein flow to liver = 185 cc/min. Symbols as
in figures 1 and 5.
tery and portal vein pressures. Epinephrine
also elevated "stump" pressure.
EFFECT OF VASOPRESSIN (PITRESSIN)
Five to ten units of vasopressin (Pitressin)
were injected into the portal vein of dogs 11,
13, 14, 15, and 16, causing a decrease of portal vein resistance and only a slight increase
of hepatic artery resistance. It was observed
also that systemic pressure rose and portal
flow decreased.
Discussion
APPRAISAL OF PERFUSION TECHNIQUE
Table 2 shows that systemic blood pressures were lower than normal in most dogs
and that some flows were below normal. Two
factors which may have caused some of the
hemodynamic parameters to be below normal
are surgical trauma and blood loss. The procedures described above required consider-
able dissection around the portal vein and the
hepatic artery. The duration of exposure of
the abdominal viscera, together with manipulation and traction of various structures
around the liver are believed to depress circulatory function in the dog.8'10 The second
factor, blood loss, was very important. In addition to unavoidable bleeding as a result of
dissection, about 500 cc of blood were required to prime the extracorporeal circuits.
In order to minimize the effects of this blood
loss, isotonic saline was injected intravenously. No other fluid or other blood was administered, because of the known adverse
effect of blood transfusions.9 Hematocrits
were not measured. Hence possible hemodilution and its effects cannot be evaluated. Since results in the dogs with poor
circulatory function were qualitatively similar
to the results in those with better function,
they were retained and reported here. The
liver of dog 11 became dark blue, swollen,
and tense, suggestive of "outflow block." With
this exception, the livers were reddish and
soft. As shown by the results, the technique
offered a method of controlling and studying
the hemodynamic parameters of the liver. Because all but one of the livers remained soft
and because the results were consistent, it is
felt that these results are valid and worthy of
analysis.
MAGNITUDE OF FLOW RATES
The average hepatic artery flow rate in
Series I was 43 cc/min/100 g of liver at 100
mm Hg. This average flow rate is perhaps
more accurate than the results of Series II because the portal vein circulation was left
intact, less surgery was required, and the circulatory status as judged by mean arterial
blood pressure was better. Hepatic arterial
flow in Series II was slightly lower for a given
perfusion pressure. This can be demonstrated
by reducing the flow data in Series II to
cc/min/100 g of liver and by plotting these
data on the graph of the Series I hepatic artery pressure-flow curves in figure 4. Generally the Series II pressure-flow points fell to
the right of the Series I pressure-flow curves,
Circulation Rtstircb, Vol. XV, Stpttmhtr 1964
HEPATIC HEMODYNAMICS
227
indicating a slightly higher hepatic artery vascular resistance in Series II.
When the average hepatic artery flow rate
noted in Series I is compared, in table 3, to
the average rates of other investigators, it will
be noted that our flow rate was greater for a
given perfusion pressure. Or to express this
another way, the vascular resistance of the
hepatic artery bed seemed to be lower in our
experiments. To explain this variation in results, some of the factors involved in measuring hepatic artery flow and our method of
handling these factors will be discussed.
Assuming that the numerous pressure and
flow measurements were accurate, these variations indicate marked differences in hepatic
arterial resistance. We noted that this resistance could be increased by hepatic nerve
stimulation. Thus, one factor tending to produce differences in resistance might be the
degree of denervation produced by surgical
trauma. In the dissection prior to cannulations
some manipulation' of nerves was unavoidable. To determine the degree of nerve dam-
TABLE 3
Downloaded from http://circres.ahajournals.org/ by guest on June 17, 2017
Hepatic Artery Blood Flow: Summary of Results Obtained by Different Methods and Investigators
Author
•nd year
Method
Anesthesia
Average BP*
Average blood flow*
mm Hg
cc/min/ 100 g
liver wt
cc/min/kg
body wt
BurtonOpifcz,11
1910
stromuhr
chloroform
ether
96.6
(75-116)
26.51
(19.4-35)
7.91
(5.0-11.8)
Macleod and
Pearce,7
1914
venous
outflow
ether
142
(110-175)
29.lt
13.4*
Blalock and
Mason,12
1936
venous
outflow
none
16.6t
(12.7-24.7)
e.it
(4.0-9.0)
Grindlay,
Herriclc, and
Mann,™
1941
thermostromuhr
none
Grodins,
Osborne,
Ivy, and
Goldman,14
1941
thermostromuhr
pentobarbital
Drapanas,
Schenk,
Pollack,
and Stewart,15
1960
noncannulating
electromagnetic
flowmeter
pentobarbital
135
(120-150)
Torrance,10
1961
perfusion
pentobarbital
100-110
27.2
(16.5-43.4)
6.9
(3.6-10.1)
Shoemaker,
1964
perfusion
pentobarbital
100
43
(28.5-58)
14.7
(7.2-20)
pentobarbital
142
—
23.91
(9.9-38.6)
Vol. XV, Stpumhtr
1964
7.2t
(3.0-11.6)
13.9t
(9.0-19.0)
* Ranges in parentheses.
f Values are averages of flows calculated from the data reported.
$ Values based on calculation that hepatic artery flow was 29% of total liver flow as
determined by these experimenters.7
Circulation Rtsttrcb,
2.6t
(1.5-3.7)
11.5t
(5.3-27.0)
228
Downloaded from http://circres.ahajournals.org/ by guest on June 17, 2017
age during the procedure, the hepatic nerves
were stimulated at the conclusion of the perfusion. Stimulation caused a marked rise in
hepatic artery resistance, thus indicating not
only that the nerves were capable of functioning, but also that throughout the nonstimulated state, during which the pressureflow curves were determined, the hepatic
artery vasomotor tone was relatively small
when compared to the high resistance observed during nerve stimulation. Another
factor, which was observed to lower the vascular resistance, was the injection of pentobarbital. As mentioned above, we included
only those results obtained at least fifteen
minutes after the last injection of anesthetic.
A third factor which might have altered the
resistance was surgical trauma. It is known
that if surgical trauma or blood loss is excessive, the liver will develop "outflow block"
and become bluish and tense with portal hypertension.8"10 It was noted that, if the liver
became cyanotic, the hepatic artery resistance
increased. Since our results did not include
those dogs in which the liver developed signs
of "outflow block," it follows that our reported average resistance would be lower
than the average resistance based on all of
our perfusions. In our opinion, these criteria
for selecting data yielded the most accurate
estimate of hepatic artery flow in the anesthetized dog; obviously this does not necessarily correlate with the hemodynamics of the
normal dog.
Another factor, which might account for
the discrepancy between our results and those
of others, relates to measurement of hepatic
artery flow. Because we noted relatively more
flow for a given perfusion pressure than did
others, the possibility of leakage in our perfusion circuit was considered. The perfusion
pump was tested to see whether increased
pressure in the outflow cannula might decrease output of the pump, and it was found
that the pump maintained calibrated flow until the perfusion pressure exceeded 250 mm
Hg. Leakage by way of collaterals was controlled by mass ligation of all structures in
the gastrohepatic ligament except for the
SHOEMAKER
hepatic artery, the portal vein, and the nerves
of the hepatic artery. Absence of leakage in
the perfusion cannula or at the site of cannulation was indicated by a relatively dry abdominal cavity after two hours of perfusion.
Measurements of perfusion pressure have
been done differently by various experimenters and may account for some variation in
results. The results shown in figure 3 indicate
that hepatic artery perfusion pressure measured in the perfusion cannula was at least two
times greater than the pressure within the
hepatic artery distal to the cannula. This
demonstration, that a significant gradient may
be present between the pressures in the perfusion cannula and the artery, indicates that
a measurement of hepatic artery resistance
based on perfusion pressure measured in the
perfusion cannula, rather than in the hepatic
artery itself, may yield lower rates of hepatic
artery flow for given perfusion pressures
owing to the artificial resistance of the cannula. Likewise, if a flow measuring device,
which distorted the artery, were interposed
between the point of pressure determination
and the liver, the pressure recorded would
not be the actual perfusion pressure. This
study emphasizes the necessity of measuring
the perfusion pressure accurately and suggests that some of the variations in measurements of hepatic artery flow rates may be related not so much to inaccuracies in measuring
blood flow, but rather to variations in methods of measuring the true perfusion pressure.
Some of the previous estimations of hepatic
artery flow indicated tiiat the hepatic artery
supplied approximately 30% of the total liver
blood flow.7' "•12 Many present day physiology texts state the same figure. Our results,
which showed a relatively higher hepatic
blood flow than those of other investigators,
suggest that hepatic artery flow may contribute a greater percentage of liver blood
flow than originally believed. The results in
Series II indicate that the hepatic artery flow
was approximately 50% of the portal flow
when the hepatic artery perfusion pressure
was only half the mean arterial blood pressure. If hepatic artery flow were to be inCircmUtiom Rtsttrcb, Vol. XV, Stpumhtr
1964
229
HEPATIC HEMODYNAMICS
resistance. Another explanation for this rise in
"stump" pressure with increased hepatic artery flow might be that all or part of the total
hepatic artery flow joins the portal flow
within the liver and that there is a detectable
vascular resistance beyond the point where
the hepatic artery and portal vein join. This
resistance, which caused the rise in "stump"
pressure during increased hepatic artery flow,
will hereafter be referred to as the "postjunctionar resistance of the portal vein. The
meanings of "pre-" and "postjunctional" resistances are indicated schematically in figure
11 where the vascular resistances are represented as variable electrical resistances. The
term "prejunctional" resistance will be applied to the vascular resistance of the portal
vein that may exist between the point where
portal vein pressure was measured and the
junction of the hepatic artery and portal
vein. The term "postjunctionar will mean
that vascular resistance between the vena
cava and the junction of the hepatic artery
and portal vein.
creased to a level such that the hepatic artery
perfusion pressure equalled systemic pressure, the hepatic artery flow could nearly
equal or exceed portal vein flow. Several investigators have noted, on occasion, that the
contribution of the hepatic artery exceeded
that of the portal vein.18-17
EFFECT OF ALTERED HEPATIC ARTERY FLOW
Downloaded from http://circres.ahajournals.org/ by guest on June 17, 2017
Increased hepatic artery blood flow increased portal vein pressure as shown in figures 3 and 5. In those dogs in which the portal vein was left intact, altered hepatic artery
flow had only a slight effect on portal vein
pressure. In the case in which both vessels
were perfused and the portal flow was held
constant, the rise in portal vein perfusion
pressure noted with increased hepatic artery
flow indicates that hepatic artery flow slightly
increases intrahepatic portal vein vascular resistance. In addition it will be noted in figure
5 that when portal flow was shunted to the
femoral vein, increased hepatic artery flow
also increased "stump" pressure. One explanation for this rise in resistance might be that
increased hepatic artery flow caused congestion, which produced increased portal vein
H
EFFECT OF ALTERED PORTAL VEIN FLOW
In Series I, acute ligation of the portal vein
A PRESS
MES ARTE RY PRESS
HE PAT I C
ART
R ESISTANCE
P O S T/-JUNCTIONAL"
R ESISTANCE
0.-0 1 P R U
MES
ART
R ESI S T A N C E
PRE-JUNCTIONAl
RESISTANCE
\
i vc
JUNCTIONAL
PRESS
PRESS
L I V E R
FIGURE 11
Electrical analog of hepatic circulation using electrical symbols to represent vascular resistances indicated in peripheral resistance units (PRU) of mm Hg/cc/min.
Circulation Resttrcb, Vol. XV, Stfinmtrr 1964
230
Downloaded from http://circres.ahajournals.org/ by guest on June 17, 2017
caused a slight decrease in hepatic artery
perfusion pressure and vascular resistance
since the hepatic artery flow was held constant. Figure 7 shows that gradual alteration
of the portal vein flow caused a direct alteration of the hepatic artery perfusion pressure
and vascular resistance, since again the arterial flow was constant. Although these results demonstrate an interrelationship between the two vascular beds, they must be
qualified by the results shown in figure 6 in
which the effects of this interrelationship
were slight. This graph shows that the hepatic
artery pressure-flow curve was not altered
markedly when the portal flow was reduced
from 190 cc/min to zero by shunting the
portal flow to the femoral vein. In summary,
the effect of altered hepatic artery flow or
portal flow was to increase the vascular resistance of the other vessel, but only slightly.
These conclusions agree qualitatively with
some of the results of Herrick who perfused
both normal and cirrhotic human livers.18
His results with normal livers were similar to
ours; namely, that in normal livers changes
in hepatic artery flow had only a small effect
on portal vein vascular resistance. He found,
however, in cirrhotic livers that the effect of
altered hepatic artery flow was more marked.
This finding of a slight interdependence of
hepatic artery and portal vein flows might be
explained by either congestion within the
liver or a common "postjunctional" vascular
resistance or a combination of both factors.
PRESSURE-FLOW CURVES
The hepatic artery pressure-flow curves
noted in Series I and shown in figure 4 were
generally linear with a slope greater than
one, and in the two experiments in which
hepatic artery flow was reduced to zero the
hepatic artery pressure remained slightly
above portal vein pressure. These results indicate that as the flow rates approached zero
the calculated vascular resistance increased,
and at higher flow rates the calculated vascular resistance decreased with increased flow
rates in five of the six dogs studied. Similar
observations were noted in Series II experi-
SHOEMAKER
ments when the hepatic artery flow was altered with fixed portal vein flow. The linearity
of the pressure-flow curves in figure 4 indicates that the hepatic artery did not show
"autoregulation" in the ranges studied.
Pressure-flow studies of the portal vein
also revealed passive vasodilatation at higher
flow rates. In figure 7 it will be noted that
portal flow increased from 75 to 450 cc/min,
but the gradient from portal vein to vena
cava increased only from 3 to 10 mm Hg,
indicating vasodilatation with increased flow
rates. This observation, that large increases in
portal vein flow resulted in small changes of
portal pressure, suggests that in the normal
dog marked changes of portal pressure are
due primarily to changes of intrahepatic vascular resistance, rather than alterations of
portal flow. The passive dilatation noted in
both the hepatic artery and the portal vein
vasculature, due to increased rates of blood
flow within the corresponding vascular bed, is
similar to the changes noted in other vascular
beds.19
Another interesting aspect of the hepatic
artery pressure-flow curves obtained with the
portal vein intact is the contrast between our
results and those of Torrance.10 Torrance used
a technique quite similar to ours, but measured the perfusion pressure with a catheter
connected to a side-arm of the perfusion cannula. In contrast to our linear pressure-flow
curves, he noted that the hepatic artery pressure-flow curve was not linear, and that for
flows above 30 cc/min/100 g liver the curve
tended to parallel the pressure axis. This nonlinearity probably was not caused by liver
congestion, since the rise in portal vein pressure secondary to increases in hepatic artery
flow was slight when measured by us and by
Torrance. Although our results showed that
pressures measured with a catheter connected
to the perfusion cannula were higher than
hepatic artery pressure, the pressure-flow
curve constructed from our measurements
was linear. One possible explanation for the
difference between our curves and those of
Torrance may be that turbulence could have
developed within the perfusion circuit used
Circulation Rtsttrcb, Vol. XV, Stpttmber 1964
HEPATIC
HEMODYNAMICS
by Torrance and still further exaggerated the
difference between true hepatic artery pressure and the pressure recorded from a sidearm of the perfusion circuit.
SYMPATHETIC NERVE STIMULATION
Downloaded from http://circres.ahajournals.org/ by guest on June 17, 2017
Sympathetic nerve stimulation produced
vasoconstriction of the hepatic artery and portal vein vasculature during the period of
stimulation. Stimulation of the hepatic nerves
in Series I increased portal vein pressure, increased hepatic artery perfusion pressure, and
increased hepatic artery vascular resistance.
In Series II in which both vessels were perfused, splanchnic stimulation caused vasoconstriction of the vascular beds of the mesenteric artery, hepatic artery, and intrahepatic
portal vein (fig. 9). The rise in systemic
pressures noted with splanchnic stimulation,
together with the initial mesenteric vasoconstriction and secondary vasodilatation, caused
an initial decrease in mesenteric flow followed
by a rebound to above normal flow rate. The
biphasic change in mesenteric flow and vascular resistance, and also the rise in hepatic
artery vascular resistance, have been noted
previously.2'8 In this same series, hepatic
nerve stimulation caused vasoconstriction of
only the hepatic artery and the vascular bed
of the intrahepatic portal vein.
As mentioned above, portal vein pressure
depends upon vena cava pressure, portal
flow, and the intrahepatic vascular resistance
to portal flow. The results in figure 8, which
show that pressure in the vena cava was constant and that mesenteric flow decreased initially, suggest that in the intact dog the initial
rise in portal pressure with splanchnic stimulation was due to increased intrahepatic portal vein vascular resistance (fig. 9). Likewise,
calculation of portal vein vascular resistance
from data shown in figure 10 indicates that
hepatic nerve stimulation produces the same
result. These data indicate that the rise in
portal pressure noted in the intact dog as a
result of sympathetic stimulation is due
probably to increased portal vein resistance.
PORTAL VEIN RESISTANCE
From the previous discussion, it will be
CircmUtion Resetrcb, Vol. XV, Stpttmber 1964
231
noted that increased hepatic artery blood
flow increased portal vein resistance, but this
does not apply to the situation of nerve stimulation. In our experiments the hepatic artery
flow was maintained constant during nerve
stimulation and, nevertheless, portal vein vascular resistance rose. Measurements in the intact dog show that hepatic artery flow decreases with this stimulation.2 Therefore the
rise in portal pressure noted with splanchnic
stimulation is probably not a result of increased hepatic artery flow.
It is concluded that there was an active vascular resistance in the liver between the portal vein and the vena cava, namely either the
"prejunctional" and/or the "postjunctionar
resistances (fig. 11). In the experiment shown
in figure 8, portal flow was shunted to the
femoral vein, and the splanchnic nerve was
restimulated. "Stump" pressure paralleled the
changes of portal vein pressure noted with
stimulation before the shunt. This indicated
that there was active vasoconstriction in the
"postjunctionar resistance. Efforts to quantitate the change in "prejunctionar resistance
were unsuccessful because of the small magnitude of this resistance; consequently it
could not be determined whether or not there
was active vasoconstriction in this resistance.
Likewise, in figure 10 it will be seen that
when the hepatic nerves were stimulated
the intrahepatic hemodynamic changes were
the same as those observed during stimulation of the splanchnic nerves. These changes
were a rise in hepatic artery and portal vein
vascular resistances. Again this rise in portal
vein resistance was shown to be due in part
to the active "postjunctjonal" vasoconstriction.
These results do not indicate what anatomical structure constitutes this "postjunctionar resistance. Distal to the junction of
the portal vein and hepatic artery there are
sinusoids and the hepatic veins with their
"sphincters."9' ^ This hepatic vein "sphincter"
mechanism has been shown to be active under
many conditions.9- 2° Evidence, such as the decrease in liver volume during sympathetic
stimulation,0 would suggest, however, that
such stimulation relaxes these "sphincters."9-20
232
Downloaded from http://circres.ahajournals.org/ by guest on June 17, 2017
The sinusoids of frogs and rats, on the other
hand, have been observed to constrict actively
with sympathetic stimulation.4'5 These sinusoids are numerous and probably offer
little vascular resistance when relaxed, but it
is conceivable that if there were sufficient
constriction of these sinusoids, there might be
an increase in the vascular resistance of the
portal vein bed. It should be recalled that the
portal system is a low pressure system. In
these experiments the vascular portal resistance had a magnitude of about 0.01 PRU,
whereas the resistance of the hepatic artery
was 1 PBU. Consequently, since the portal
system is such a low resistance system, constriction of the sinusoids might result in a
slight absolute increase in resistance, which
would be a significant increase. It also has
been shown in dogs that dye injected into the
portal vein passes through the liver more rapidly during sympathetic stimulation.21 In fact,
however, it remains to be determined what
constitutes this active "postjunctionai" resistance. The rise in portal vein pressure with
splanchnic or hepatic nerve stimulation in the
intact dog is believed to be due in part to
active vasoconstriction of the "postjunctional" resistance.
SECONDARY VASODILATATION
It was shown that sympathetic stimulation
caused an initial vasoconstriction. If the control vasomotor tone was high, these vascular
beds showed a secondary vasodilatation. Our
study provides no explanation for this finding.
EFFECT OF DRUGS
The rise in both hepatic artery and portal
vein resistances with injection of epinephrine
was due in part to constriction of the "postjunctional" resistance, as with sympathetic
stimulation. The role of the "prejunctional"
resistance could not be evaluated.
Pitressin injected into the portal vein
caused a slight increase in hepatic artery resistance, whereas the portal vein resistance
decreased as reported by Butz et al.22
Bute and his group have pointed out that this
portal vein vasodilatation and the known decreased portal flow with Pitressin both tend
SHOEMAKER
to lower portal pressure.22 It has been shown
previously that Pitressin probably increases
hepatic artery flow in the intact dog.23 Our
results, consisting of an increased systemic
blood pressure with only a slight increase in
hepatic artery resistance, would lead to the
same conclusion; namely, Pitressin can be expected to increase hepatic artery flow in the
intact animal.
Summary
The effects of altered hepatic blood flow, of
splanchnic and hepatic nerve stimulation, and
of injecting epinephrine and vasopressin
(Pitressin) were studied by perfusing canine
livers in vitro. Controlled hepatic arterial perfusions demonstrated a linear relationship between arterial pressure and hepatic arterial
inflow. It was found that increases of hepatic
artery flow slightly increased portal vein pressure and vascular resistance of the portal vein,
and that increases of portal vein flow slightly
increased hepatic artery vascular resistance.
Splanchnic nerve stimulation caused vasoconstriction of the hepatic artery, portal vein, and
mesenteric vasculature. Hepatic nerve stimulation caused constriction of only the hepatic
artery and portal vein vasculature. Similar results were obtained by the administration of
epinephrine. Pitressin caused marked vasoconstriction of the mesenteric vasculature,
vasodilatation of the portal vein vascular bed,
and slightly increased hepatic artery vasomotor tone. It was noted that the rise in portal
vein vasomotor tone with sympathetic stimulation was due in part to constriction of the
"postjunctional" resistance.
Acknowledgment
The author expresses his gratitude to Drs. S.
R. Powers and R. S. Alexander and to the many
others who have made this work possible.
References
1.
BAYLJSS, W. M., AND STABLING, E. H.:
Observa-
tions on venous pressure and their relationship to capillary pressures. J. PhysioL 16: 159,
1894.
2.
GREEN, H.
D.,
HALL, L.
S.,
SEXTON, J.,
AND
DEAL, C. P.: Autonomic vasomotor responses
in the canine hepatic arterial and venous
beds. Am. I. Physiol. 196: 196, 1959.
CircmlMlion Research, Vol. XV, September 1964
233
HEPATIC HEMODYNAMICS
3.
C : Measurement of the blood flow of the
liver. Am. J Physiol. 132: 489, 1941.
DEAL, C. P., AND GREEN, H. D.: Comparison of
changes in mesenteric resistance following
splanchnic nerve stimulation with responses to
epinephrine and norepinephrine. Circulation
Res. 4: 38, 1956.
4. SENEVIHATNE, R. D.: Physiological and pathological responses in the blood-vessels of the liver.
Quart. J. Exptl. Physiol. 35: 77, 1949-50.
5. WAKIM, K. G.: Effects of stimulation of autonomic ner\'es on intrahepatic circulation of
blood in intact animal. Proc. Soc. Exptl. Biol.
Med. 49: 307, 1942.
E. L., AND STEWART, J. D.: Hepatic hemodynamics in experimental ascites. Ann. Surg.
152: 705, 1960.
16. TORRANCE, H. B.: The control of the hepatic
arterial circulation. J. Physiol. 158: 39, 1961.
6.
17.
GRIFFITH, F. R., JR., AND EMERY, F. E.: The
vasomotor control of the liver circulation Am.
J. Physiol. 95: 20, 1930.
7. MACLEOD, J. J. R., AND PEARCE, R. G.: The out-
Downloaded from http://circres.ahajournals.org/ by guest on June 17, 2017
flow of blood from the liver as affected by
variations in the condition of the portal vein
and hepatic artery. Am. J. Physiol. 35: 87,
1914.
8. KESTENS, P. L., AND MCDERMOTT, W. V.: Per-
fusion and replacement of the canine liver.
Surgery 50: 196, 1961.
9. ANDREWS, W. H. H., HECKER, R., MAECRAITH,
B. G., AND RITCHIE, H. D.: The action of
adrenaline, Z-noradrenaline, acetycholine, and
other substances on the blood vessels of the
perfused canine liver. J. Physiol. 128: 413,
1955.
10.
13.
15.
GRINDLAY, J. H., HEHRICK, J. F., AND MANN, F.
Circulation Reietrcb, Vol. XV, Sepumber 1964
DRAPANAS, T., SCHENK,
W. G., JR., POLLACK,
SOSKIN, S., ESSEX, H. E., HERRICK, J. F., AND
MANN, F. C.: The mechanism of regulation
of the blood sugar by the liver. Am. J. Physiol.
124: 558, 1938.
18. HERRICK, F. C : An experimental study into the
cause of the increased portal pressure in
portal cirrhosis. J. Exptl. Med. 9: 93, 1907.
19. RUCH, T. C , AND FULTON, J. F.: Medical Physi-
ology and Biophysics. Philadelphia, W. B.
Saunders Company, 1960, p. 659.
20. DEYSACH, L. J.: The nature and location of the
"sphincter mechanism" in the liver as determined by drug actions and vascular injections.
Am. J. Physiol. 132: 713, 1941.
21.
DANIEL, P. M., AND PRICHARD, M. M. L.: Effects
of stimulation of the hepatic nerves and of
adrenaline upon the circulation of the portal
venous blood within the liver. J. Physiol. 114:
538, 1951.
22.
BUTZ, G. W., JR., LAZARUS, R. E., SCHILLER,
W. R., and STARZL, T. E.: Studies of the
effects and modes of action of autonomic
drugs on portal hemodynamics. Surgery 51:
364, 1962.
BLALOCK, A., AND MASON, M. F.: Observations
on the blood flow and gaseous metabolism of
the liver of unanaesthetised dogs. Am. J.
Physiol. 117: 328, 1936.
GRODINS, F. S., OSBORNE, S. L., IVY, A. C., AND
GOLDMAN, L.: The effect of bile acids on
hepatic blood flow. Am. J. Physiol. 132: 375,
1941.
SHOEMAKER, W. C., PANICO, F. G., WALKER, W.
F., AND ELWYN, D. H.: Perfusion of canine
liver in vivo. J. Appl. Physiol. 15: 687, 1960.
11. BURTON-OPITZ, R.: The vascularity of the liver.
I. The flow of blood in the hepatic artery.
Quart. J. Exptl. Physiol. 3: 297, 1910.
12.
14.
23.
HEIMBURCER,
I.,
TERAMOTO,
S.,
AND SHU-
MACKER, H. B., JR.: Effect of surgical pituitrin
upon portal and hepatic circulation. Surgery
48: 706, 1960.
A Study of Hepatic Hemodynamics in the Dog
CHARLES P. SHOEMAKER, Jr.
Downloaded from http://circres.ahajournals.org/ by guest on June 17, 2017
Circ Res. 1964;15:216-233
doi: 10.1161/01.RES.15.3.216
Circulation Research is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
Copyright © 1964 American Heart Association, Inc. All rights reserved.
Print ISSN: 0009-7330. Online ISSN: 1524-4571
The online version of this article, along with updated information and services, is located on the
World Wide Web at:
http://circres.ahajournals.org/content/15/3/216
Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published in
Circulation Research can be obtained via RightsLink, a service of the Copyright Clearance Center, not the
Editorial Office. Once the online version of the published article for which permission is being requested is
located, click Request Permissions in the middle column of the Web page under Services. Further information
about this process is available in the Permissions and Rights Question and Answer document.
Reprints: Information about reprints can be found online at:
http://www.lww.com/reprints
Subscriptions: Information about subscribing to Circulation Research is online at:
http://circres.ahajournals.org//subscriptions/