AMERICAN
JOURNAL
OF hIYSIOLOGY
Vol. 230, No. 5, May 1976. Printed
in USA.
Direct effects of various catecholamines liver circulation in dogs
on
LEROY J. HIRSCH, TAKAO AYABE, AND GERALD GLICK Cardiovascular Institute, Department of Medicine, Michael Reese Hospital and Medical and University of Chicago Pritzker School of Medicine, Chicago, Illinois 60616
Center,
HIRSCH,LEROY J.,TAKAO AYABE, AND GERALD GLICK. Direct effects of various catecholamines on liver circulation in dogs. Am. J. Physiol. 230(5): 1394-1399. 1976. -As measured by electromagnetic blood flow transducers, direct infusion of epinephrine, norepinephrine, and dopamine into the portal vein (PV) produced a 40-50% decrease in hepatic arterial (HA) blood flow; isoproterenol increased HA flow by about 69%. No changes in PV flow or pressure were observed. Direct HA infusion of the vasoconstrictors decreased HA flow by amounts comparable to those occurring after PV infusion. However, HA infusion of isoproterenol increased HA flow only 15%, suggesting a difference in beta-receptor population in the two vessels. When infused directly into the superior mesenteric artery @MA), epinephrine and norepinephrine reduced SMA flow by about 45% and PV flow by 20-25%; HA flow increased 64%. Infusion of isoproterenol and dopamine into SMA increased SMA flow by 115% and 206% and PV flow by 60% and 70%, respectively, whereas HA flow decreased by 25% and 50%. Portal vein pressure increased less than 3 mmHg. Alphaand beta-receptor blockade of the liver did not change signif!cantly the alterations in hepatic arterial blood flow that were secondary to changes in portal venous blood flow. It is likely that regulation of hepatic arterial flow resides in mechanisms located within the liver sinusoids.
fairly well known (6), this phenomenon has not generally been appreciated during exogenous administration of pharmacologic agents. Many vasoactive agents are administered with little regard for their effect on organ systems other than those that are of immediate concern. Ross and Brown (15) have described a vasoconstrictor effect of dopamine on hepatic arterial flow when administered intravenously, while simultaneously observing a marked vasodilation in the mesenteric bed. Similarly, Shanbour and Jacobson (18) and others (17) have shown that cardiac glycosides produce a significant effect on splanchnic and peripheral vascular resistance in addition to their cardiac effect. Since the splanchnic bed receives as much as 25% of the total cardiac output, and it is known to be a reservoir for circulating blood (2), we designed the present study to elucidate the response of the splanchnic bed in general, and the hepatic bed in particular, to pharmacologic agents that are generally available for therapeutic purposes. We also have tried to define the roles of various receptors in these responses.
portal blood flow; hepatic arterial blood flow; total liver blood flow; isoproterenol; norepinephrine; epinephrine; dopamine; alpha- and beta-adrenergic blockade; autoregulation
METHODS
OF VASOACTIVE RECEPTOR SITES in the vascular bed of the liver has been reported by numerous investigators (3, 16). Alpha and beta receptors are known to exist in the mesenteric bed (8). Although alpha receptors have been reported to be present in the hepatic arterial and portal venous vessels and beta receptors appear to be present in the hepatic arterial bed (3, 4, 7, 16), beta receptors have not been shown to be of much, if any, significance in the portal vessels (3, 8). Indeed, Hanson (8) has indicated recently that intraportal infusion of isoproterenol did not alter flow or resistance in the portal vein. Furthermore, dopaminergic vasodilator receptors unrelated to beta receptors have been demonstrated in the renal and mesenteric beds (15). The va soconstrictor effect of dopamine has been ascribed to stimulation of alpha receptors (12, 13). However, little attention has been directed toward the effect of dopamine on hepatic hemodynamics. Although the autoregulatory reciprocal relation between the portal venous flow and hepatic arterial flow is THE PRESENCE
Twenty-five dogs weighing between 17 and 20 kg were anesthetized with pentobarbital sodium (25 mg/kg). The animals were intubated to maintain a free airway. A flank incision was made just below the right costal margin. The superior mesenteric artery, the common hepatic artery, and the portal vein were exposed by blunt dissection, with care being taken to maintain as much of their neural integrity as possible. Electromagnetic bloodflow transducers (Biotronex Lab., Inc., Silver Spring, Md.) were placed around each vessel such that a tight fit was maintained. Hydraulic occluders (Rhodes Medical Instruments, Woodland Hills, Calif.) were placed downstream from the probes. Any vessels between the probes and the occluders were ligated and cut. Periodically during the experiments the vessels were occluded to ascertain zero flow and to detect any drift. Polyethylene tubing (PE-90) was inserted into each vessel by way of side branches. The tubing was used for pressure measurements and for the infusion of the several drugs. Since the portal vein catheter was used for both infusion and pressure measurement, portal vein pressure could not be measured during portal infusion. The common hepatic artery was ligated distal to the
1394
Downloaded from www.physiology.org/journal/ajplegacy by ${individualUser.givenNames} ${individualUser.surname} (129.108.009.184) on October 24, 2018. Copyright © 1976 American Physiological Society. All rights reserved.
EFFECTS
OF
CATECHOLAMINES
ON
LIVER
CIRCULATION
origin of the proper hepatic arteries, as it becomes the gastroduodenal artery. The ligation ensured measurement of total arterial inflow to the liver (Fig. 1). Under fluoroscopic control, catheters were also placed at the root of the aorta and in a hepatic vein for pressure measurement. The following vasoactive drugs were infused into each vessel at the indicated doses that, as determined previously in pilot experiments in other dogs, produced the maximum flow response without eliciting systemic effects, i.e., alterations in heart rate or arterial pressure: isoproterenol 0.1 pug/kg per min, norepinephrine 0.1 PgI kg per min, epinephrine 0.1 Fg/kg per min, and dopamine 2.0 pg/kg per min. The drugs were diluted in normal saline so that each milliliter contained the required concentration per kilogram and was infused at a rate of 1.0 ml/min with a Harvard infusion pump. The concentrations of the active agents were calculated as the amount of free base infused. The drugs were infused into each vessel in random order with an interval of at least 10 min between infusions. The catheters were always cleared with a 3.0-ml flush of saline at the end of an infusion. A total of three infusions per vessel per drug was the usual protocol for each experiment. After the initial series of infusions, alpha- and betareceptor blocking agents (1 mg phenoxybenzamine and 1 mg propranolol) were infused directly into either the superior mesenteric artery or the portal vein. We allowed 1 h for phenoxybenzamine to exert its alphablocking effect and 15 min for propranolol to exert its beta-blocking effect, and then the vasoactive drugs were infused once again. The significance of differences in mean values was tested by the two-tailed Student t test for paired data. RESULTS
Isoproterenol. Isoproterenol, 0.1 pg/kg per min, when infused directly into the portal vein, consistently produced a marked increase in hepatic arterial flow (Fig. 2A, Table l), the average increment being 69 -+ 7% above control values. This large increase in hepatic arterial flow produced by intraportal infusion conCOMMON L
HEPATIC
A
GA PROPER
HEPATIC
GASTRODUOOENAL
R
GASTRIC
PANCREATICODUODENAL
I T-
PORTAL
MESENTERIC
U
A A
A
V
UJ,\
1. Diagram showing relationship transducers (EMF) and hydraulic occluders portal venous systems. FIG.
A
A
R GAGTROEPIPLOIC
SPLENIC
AA
of electromagnetic to hepatic arterial
blood and
EPI
NE
DOP
20
y
1
B
‘-: C
FIG. 2. Response of superior mesenteric arterial flow (SMA), portal venous flow (PV), and hepatic arterial flow (HA) to direct portal venous infusion of: A, isoproterenol 0.1 Kg/kg per min; B, epinephrine 0.1 pg/kg per min; C, norepinephrine 0.1 @g/kg per min; and 0, dopamine 2.0 pg/kg per min.
trasted with only a 15 ? 5% increase in hepatic arterial flow when isoproterenol was infused directly into the hepatic artery itself (Fig. 3A, Table 1). No significant changes were produced in portal venous flow or hepatic venous pressure by infusions of isoproterenol into the portal vein or hepatic artery. Little or no changes in heart rate or aortic pressure were detected, an indication that the isoproterenol did not escape in significant amounts into the general circulation. When infused into the superior mesenteric artery (Fig. 4A), isoproterenol produced an increase in superior mesenteric arterial flow of 115 t 8% while hepatic arterial flow decreased about 25%. Portal venous pressure, as a consequence of a 60% increase in flow through the portal vein, rose by as much as 2 mmHg and hepatic venous pressure also increased as much as 1 mmHg (Table 1). Epinephrine and norepinephrine. Epinephrine and norepinephrine, 0.1 pg/kg per min, when infused into the superior mesenteric artery, produced a 40-50% reduction in superior mesenteric arterial flow; as a consequence the portal flow fell approximately 20-25% (Fig. 4, B and C) and portal pressure decreased by almost 1 mmHg (Table 1). Hbpatic arterial flow increased by 68% and hepatic venous pressure remained relatively unchanged. When infused into the portal vein (Fig. 2, B and C) or hepatic artery (Fig. 3, B and C), the two catecholamines reduced hepatic arterial flow by as much as 50% with only minor alterations in hepatic venous pressure. Changes in portal venous flow did not occur with infusion into either the portal vein or hepatic artery. Unlike isoproterenol, which produced a greater effect when infused into the portal vein compared to the hepatic artery, epinephrine and norepinephrine caused comparable reductions in hepatic arterial flow as a result of infusion into either of these vessels. Dopamine. Dopamine, 2.0 pglkg per min, infused directly into the superior mesenteric artery (Figs. 40 and 5), produced in this vessel an initial short-lived vasoconstriction with reduction in flow and then a pronounced increase in flow that eventually stabilized at levels in excess of 200% of control values for the duration of the infusion (4-5 min). Portal venous flow increased
Downloaded from www.physiology.org/journal/ajplegacy by ${individualUser.givenNames} ${individualUser.surname} (129.108.009.184) on October 24, 2018. Copyright © 1976 American Physiological Society. All rights reserved.
1396
HIRSCH,
TABLE 1. Effects of catecholamines and hepatic arterial blood flows Site of Infusion
Dw3 Isoproterenol, 0.1 pglkg per min
Epinephrine, 0.1 pg/kg per min
Norepinephrine, 0.1 pg/kg per min
Dopamine, 2.0 &kg per min
on superior
SMAQ Control
mesenteric
arterial,
HAG
Exptl
P
Control
portal
PVP
Exptl
P
Control
Exptl
P
Control
Exptl
P
7.020.4
8.320.3
co.01
4.6kO.l 4.420.2 4.820.1
5.350.1 4.520.3 4.8kO.l
co.01 NS NS
6.520.7 5.320.4 4.8kO.3
6.220.8 5.2kO.3 5.OkO.3
NS NS NS
5.620.2 5.220.4 6.420.2
5.820.3 5.120.2 6.520.3
NS NS NS
3.3kO.2 4.5kO.4 5.320.2
4.020.2 4.220.5 5.620.4
co.01 NS NS
533241 586228 602250
870+28 605250 590535
JOURNAL
OF hIYSIOLOGY
Vol. 230, No. 5, May 1976. Printed
in USA.
Direct effects of various catecholamines liver circulation in dogs
on
LEROY J. HIRSCH, TAKAO AYABE, AND GERALD GLICK Cardiovascular Institute, Department of Medicine, Michael Reese Hospital and Medical and University of Chicago Pritzker School of Medicine, Chicago, Illinois 60616
Center,
HIRSCH,LEROY J.,TAKAO AYABE, AND GERALD GLICK. Direct effects of various catecholamines on liver circulation in dogs. Am. J. Physiol. 230(5): 1394-1399. 1976. -As measured by electromagnetic blood flow transducers, direct infusion of epinephrine, norepinephrine, and dopamine into the portal vein (PV) produced a 40-50% decrease in hepatic arterial (HA) blood flow; isoproterenol increased HA flow by about 69%. No changes in PV flow or pressure were observed. Direct HA infusion of the vasoconstrictors decreased HA flow by amounts comparable to those occurring after PV infusion. However, HA infusion of isoproterenol increased HA flow only 15%, suggesting a difference in beta-receptor population in the two vessels. When infused directly into the superior mesenteric artery @MA), epinephrine and norepinephrine reduced SMA flow by about 45% and PV flow by 20-25%; HA flow increased 64%. Infusion of isoproterenol and dopamine into SMA increased SMA flow by 115% and 206% and PV flow by 60% and 70%, respectively, whereas HA flow decreased by 25% and 50%. Portal vein pressure increased less than 3 mmHg. Alphaand beta-receptor blockade of the liver did not change signif!cantly the alterations in hepatic arterial blood flow that were secondary to changes in portal venous blood flow. It is likely that regulation of hepatic arterial flow resides in mechanisms located within the liver sinusoids.
fairly well known (6), this phenomenon has not generally been appreciated during exogenous administration of pharmacologic agents. Many vasoactive agents are administered with little regard for their effect on organ systems other than those that are of immediate concern. Ross and Brown (15) have described a vasoconstrictor effect of dopamine on hepatic arterial flow when administered intravenously, while simultaneously observing a marked vasodilation in the mesenteric bed. Similarly, Shanbour and Jacobson (18) and others (17) have shown that cardiac glycosides produce a significant effect on splanchnic and peripheral vascular resistance in addition to their cardiac effect. Since the splanchnic bed receives as much as 25% of the total cardiac output, and it is known to be a reservoir for circulating blood (2), we designed the present study to elucidate the response of the splanchnic bed in general, and the hepatic bed in particular, to pharmacologic agents that are generally available for therapeutic purposes. We also have tried to define the roles of various receptors in these responses.
portal blood flow; hepatic arterial blood flow; total liver blood flow; isoproterenol; norepinephrine; epinephrine; dopamine; alpha- and beta-adrenergic blockade; autoregulation
METHODS
OF VASOACTIVE RECEPTOR SITES in the vascular bed of the liver has been reported by numerous investigators (3, 16). Alpha and beta receptors are known to exist in the mesenteric bed (8). Although alpha receptors have been reported to be present in the hepatic arterial and portal venous vessels and beta receptors appear to be present in the hepatic arterial bed (3, 4, 7, 16), beta receptors have not been shown to be of much, if any, significance in the portal vessels (3, 8). Indeed, Hanson (8) has indicated recently that intraportal infusion of isoproterenol did not alter flow or resistance in the portal vein. Furthermore, dopaminergic vasodilator receptors unrelated to beta receptors have been demonstrated in the renal and mesenteric beds (15). The va soconstrictor effect of dopamine has been ascribed to stimulation of alpha receptors (12, 13). However, little attention has been directed toward the effect of dopamine on hepatic hemodynamics. Although the autoregulatory reciprocal relation between the portal venous flow and hepatic arterial flow is THE PRESENCE
Twenty-five dogs weighing between 17 and 20 kg were anesthetized with pentobarbital sodium (25 mg/kg). The animals were intubated to maintain a free airway. A flank incision was made just below the right costal margin. The superior mesenteric artery, the common hepatic artery, and the portal vein were exposed by blunt dissection, with care being taken to maintain as much of their neural integrity as possible. Electromagnetic bloodflow transducers (Biotronex Lab., Inc., Silver Spring, Md.) were placed around each vessel such that a tight fit was maintained. Hydraulic occluders (Rhodes Medical Instruments, Woodland Hills, Calif.) were placed downstream from the probes. Any vessels between the probes and the occluders were ligated and cut. Periodically during the experiments the vessels were occluded to ascertain zero flow and to detect any drift. Polyethylene tubing (PE-90) was inserted into each vessel by way of side branches. The tubing was used for pressure measurements and for the infusion of the several drugs. Since the portal vein catheter was used for both infusion and pressure measurement, portal vein pressure could not be measured during portal infusion. The common hepatic artery was ligated distal to the
1394
Downloaded from www.physiology.org/journal/ajplegacy by ${individualUser.givenNames} ${individualUser.surname} (129.108.009.184) on October 24, 2018. Copyright © 1976 American Physiological Society. All rights reserved.
EFFECTS
OF
CATECHOLAMINES
ON
LIVER
CIRCULATION
origin of the proper hepatic arteries, as it becomes the gastroduodenal artery. The ligation ensured measurement of total arterial inflow to the liver (Fig. 1). Under fluoroscopic control, catheters were also placed at the root of the aorta and in a hepatic vein for pressure measurement. The following vasoactive drugs were infused into each vessel at the indicated doses that, as determined previously in pilot experiments in other dogs, produced the maximum flow response without eliciting systemic effects, i.e., alterations in heart rate or arterial pressure: isoproterenol 0.1 pug/kg per min, norepinephrine 0.1 PgI kg per min, epinephrine 0.1 Fg/kg per min, and dopamine 2.0 pg/kg per min. The drugs were diluted in normal saline so that each milliliter contained the required concentration per kilogram and was infused at a rate of 1.0 ml/min with a Harvard infusion pump. The concentrations of the active agents were calculated as the amount of free base infused. The drugs were infused into each vessel in random order with an interval of at least 10 min between infusions. The catheters were always cleared with a 3.0-ml flush of saline at the end of an infusion. A total of three infusions per vessel per drug was the usual protocol for each experiment. After the initial series of infusions, alpha- and betareceptor blocking agents (1 mg phenoxybenzamine and 1 mg propranolol) were infused directly into either the superior mesenteric artery or the portal vein. We allowed 1 h for phenoxybenzamine to exert its alphablocking effect and 15 min for propranolol to exert its beta-blocking effect, and then the vasoactive drugs were infused once again. The significance of differences in mean values was tested by the two-tailed Student t test for paired data. RESULTS
Isoproterenol. Isoproterenol, 0.1 pg/kg per min, when infused directly into the portal vein, consistently produced a marked increase in hepatic arterial flow (Fig. 2A, Table l), the average increment being 69 -+ 7% above control values. This large increase in hepatic arterial flow produced by intraportal infusion conCOMMON L
HEPATIC
A
GA PROPER
HEPATIC
GASTRODUOOENAL
R
GASTRIC
PANCREATICODUODENAL
I T-
PORTAL
MESENTERIC
U
A A
A
V
UJ,\
1. Diagram showing relationship transducers (EMF) and hydraulic occluders portal venous systems. FIG.
A
A
R GAGTROEPIPLOIC
SPLENIC
AA
of electromagnetic to hepatic arterial
blood and
EPI
NE
DOP
20
y
1
B
‘-: C
FIG. 2. Response of superior mesenteric arterial flow (SMA), portal venous flow (PV), and hepatic arterial flow (HA) to direct portal venous infusion of: A, isoproterenol 0.1 Kg/kg per min; B, epinephrine 0.1 pg/kg per min; C, norepinephrine 0.1 @g/kg per min; and 0, dopamine 2.0 pg/kg per min.
trasted with only a 15 ? 5% increase in hepatic arterial flow when isoproterenol was infused directly into the hepatic artery itself (Fig. 3A, Table 1). No significant changes were produced in portal venous flow or hepatic venous pressure by infusions of isoproterenol into the portal vein or hepatic artery. Little or no changes in heart rate or aortic pressure were detected, an indication that the isoproterenol did not escape in significant amounts into the general circulation. When infused into the superior mesenteric artery (Fig. 4A), isoproterenol produced an increase in superior mesenteric arterial flow of 115 t 8% while hepatic arterial flow decreased about 25%. Portal venous pressure, as a consequence of a 60% increase in flow through the portal vein, rose by as much as 2 mmHg and hepatic venous pressure also increased as much as 1 mmHg (Table 1). Epinephrine and norepinephrine. Epinephrine and norepinephrine, 0.1 pg/kg per min, when infused into the superior mesenteric artery, produced a 40-50% reduction in superior mesenteric arterial flow; as a consequence the portal flow fell approximately 20-25% (Fig. 4, B and C) and portal pressure decreased by almost 1 mmHg (Table 1). Hbpatic arterial flow increased by 68% and hepatic venous pressure remained relatively unchanged. When infused into the portal vein (Fig. 2, B and C) or hepatic artery (Fig. 3, B and C), the two catecholamines reduced hepatic arterial flow by as much as 50% with only minor alterations in hepatic venous pressure. Changes in portal venous flow did not occur with infusion into either the portal vein or hepatic artery. Unlike isoproterenol, which produced a greater effect when infused into the portal vein compared to the hepatic artery, epinephrine and norepinephrine caused comparable reductions in hepatic arterial flow as a result of infusion into either of these vessels. Dopamine. Dopamine, 2.0 pglkg per min, infused directly into the superior mesenteric artery (Figs. 40 and 5), produced in this vessel an initial short-lived vasoconstriction with reduction in flow and then a pronounced increase in flow that eventually stabilized at levels in excess of 200% of control values for the duration of the infusion (4-5 min). Portal venous flow increased
Downloaded from www.physiology.org/journal/ajplegacy by ${individualUser.givenNames} ${individualUser.surname} (129.108.009.184) on October 24, 2018. Copyright © 1976 American Physiological Society. All rights reserved.
1396
HIRSCH,
TABLE 1. Effects of catecholamines and hepatic arterial blood flows Site of Infusion
Dw3 Isoproterenol, 0.1 pglkg per min
Epinephrine, 0.1 pg/kg per min
Norepinephrine, 0.1 pg/kg per min
Dopamine, 2.0 &kg per min
on superior
SMAQ Control
mesenteric
arterial,
HAG
Exptl
P
Control
portal
PVP
Exptl
P
Control
Exptl
P
Control
Exptl
P
7.020.4
8.320.3
co.01
4.6kO.l 4.420.2 4.820.1
5.350.1 4.520.3 4.8kO.l
co.01 NS NS
6.520.7 5.320.4 4.8kO.3
6.220.8 5.2kO.3 5.OkO.3
NS NS NS
5.620.2 5.220.4 6.420.2
5.820.3 5.120.2 6.520.3
NS NS NS
3.3kO.2 4.5kO.4 5.320.2
4.020.2 4.220.5 5.620.4
co.01 NS NS
533241 586228 602250
870+28 605250 590535
