AMERICAN JOURNAL OF PHYSIOLOGY Vol. 230, No. 2, February 1976. Prin ted in U.S.A.
Primate gastric catecholamines MICHAEL Division
circulation: effects of and adrenergic blockade
J. ZINNER, of Surgery,
JOHN
Walter
C. KERR,
Reed Army
AND
Institute
ZINNER, MICHAEL J., JOHN C. KERR, AND DAVID G. REYNoms. Primate gastric circulation: effects of catecholamines and adrenergic blockade. Am. J. Physiol. 230(2): 346350. 1976. -The effects of intra-arterial injections and infusions of epinephrine, norepinephrine, and isoproterenol on gastric blood flow were studied in anesthetized baboons, Blood flow was measured electromagnetically before and after adrenergic blockade. The results for injected epinephrine and norepinephrine indicate these agents to be pure vasoconstrictors in the primate gastric circulation, and this response is attenuated by alpha-adrenergic blockade with phenoxybenzamine. Isoproterenol is a pure vasodilator, and its response is attenuated following beta-adrenergic blockade with proprano101. Intra-arterial infusions of epinephrine and norepinephrine C.05 pg kg-’ min-‘) resulted in sustained vasoconstriction with no evidence of autoregulatory escape and no postinfusion “overshoot.” This study suggests that epinephrine and norepinephrine might provide alternatives to vasopressin as a vasoconstrictor for the control of upper gastrointestinal bleeding.
epinephrine; norepinephrine; flow; autoregulatory escape
isoproterenol;
gastric
blood
andblockade on the splanchnic circulation have been studied by several methods and in several species. Previous investigators have reported that epinephrine and norepinephrine exert both vascoconstrictor and vasodilator effects on the gastric circulation. Whereas Walder (10) reported epinephrine to be a vasodilator in excised human stomachs, Thompson and Vane (9) found intravenous infusions of epinephrine, at low dose, to be a dilator of the gastric circulation and a high dose of epinephrine to cause vasoconstriction. Using the same intravenous dose range as Thompson and Vane, Cummings et al. (3) concluded that epinephrine was strictly a dilator and norepinephrine a constrictor in the canine gastric circulation. Delaney and Grim (4), using 42K clearance as a measure of blood flow in the dog, concurred with the observations that intravenous epinephrine increased blood flow and intravenous norepinephrine decreased blood flow to the stomach. In recent studies using intraarterial injections (12) and infusions (13) of catecholamines, we observed epinephrine to stimulate both alpha- and beta-adrenergic receptors in the canine gastric circulation. In this species epinephrine proved to be a dilator in the left gastric circulation. Norepinephrine had both alpha- and beta-adrenergic properties but unlike epinephrine was predominantly a vasoconstrictor. THE
EFFECTS
OF ADRENERGIC
STIMULATION
DAVID
G. REYNOLDS
of Research,
Washington,
D.C. 20012
The diversity of opinions concerning the action of these catecholamines in the gastric circulation is undoubtedly related to differences in species, routes and methods of drug delivery, blood flow monitoring techniques, and dose of drug employed. The action of this class of drugs is important in light of the clinical introduction of selective intra-arterial infusions of vasoconstrictor agents to control gastrointestinal bleeding (7). Catecholamines, either alone or in combination with beta-adrenergic blockade, have been used in all regional splanchnic circulations (8, 11). On the basis of our observations in dogs (13) that intra-arterial infusions of epinephrine into the gastric circulation result in either vasodilation or brief vasoconstriction followed by autoregulatory escape, we cautioned the clinical use of these catecholamines to control bleeding from gastric sources. The present study evaluates the effects of intra-arterial injections and infusions of epinephrine, norepinephrine, and isoproterenol into the gastric circulation of the subhuman primate before and after differential adrenergic blockade. MATERIALS
AND
METHODS
Ten adult Papio anubis baboons of either sex weighing 13.4-28.3 kg (mean 19.9 2 1.5 kg) were immobilized with phencyclidine hydrochloride (Sernylan, Bio-ceutic Laboratories, St. Joseph) (1.0 mg kg-l, im). The animals were then maintained with light pentobarbital anesthesia throughout the study. Through a midline laparotomy, the celiac artery was exposed and the transducer (In Vivo Metric Systems, Los Angeles) of a gated sinewave electromagnetic blood flowmeter (Biotronex Laboratories, Inc., Silver Spring, Md.) was placed on the artery near its origin. A hydraulic occluder was placed distally on the artery for zero flow determinations. The common hepatic artery was isolated and cannulated with a PE-90 polyethylene catheter (Clay-Adams, Inc., New York) for intra-arterial delivery of drugs. A splenectomy was performed with careful attention to the maintenance of the short gastric blood vessels (Fig. 1). Catheters were inserted into the abdominal aorta through a femoral artery (PE-190) and into the portal vein (PE-90) through a branch of the superior mesenteric vein and connected to pressure transducers (Statham Instruments, Inc., Oxnard, Calif.) for monitoring arterial and portal pressures. A femoral vein was cannulated (PE-190) for intravenous administration of drugs. Pressure transducers were calibrated prior to each experiment with a mercury manometer. Blood flow trans-
Downloaded from www.physiology.org/journal/ajplegacy by ${individualUser.givenNames} ${individualUser.surname} (130.070.008.131) on January 11, 2019.
PRIMATE
GASTRIC
347
CIRCULATION
ducers were calibrated with whole blood in vitro and lated gastric vascular resistance would be reflected by checked in vivo. All parameters were recorded simultachanges in gastric blood flow. neously on a multichannel recorder (Sanborn, WalIntra-arterial injections. The effects of intra-arterial tham). injections of epinephrine in the baboon gastric circulaThe drugs injected and infused were epinephrine hytion are seen in Fig. 2. At the lowest dose, 10e3pg kg-‘, drochloride (The Vitarine Co., New York), norepinephepinephrine elicited a decrease in flow of 0.35 t .12 ml rine bitartrate (Levophed, Winthrop Laboratories, New min-’ kg-‘. As the dose was increased, the constrictor York), and isoproterenol (Isuprel, Winthrop). Using sa- response increased to 2.90 ? .36 ml min-’ kg-’ at the line as a diluent, drugs were injected intra-arterially as highest dose. Beta-adrenergic blockade had no signifia O.l-ml bolus in doses ranging logarithmically from cant effect on the responses, but alpha-adrenergic blocklop3 to 10’ pg (base) kg-’ followed by a l.O-ml flush of ade significantly reduced the response (P < .OOl>.Intrasaline. Intra-arterial infusions of drugs were delivered arterial injections of norepinephrine over the same dose at a rate of 0.05 pg (base) kg-’ min-’ for 10 min. The range (Fig. 3) resulted in similar but somewhat smaller dose of drug employed was selected to give maximum responses. At the highest dose, 1.0 pg kg-‘, the decrease reproducible responses in total gastric blood flow within gastric blood flow was 2.42 t .39 ml min-’ kg-‘. Betaout altering arterial or portal pressures. adrenergic blockade did not significantly influence the Following the control responses, half the animals dose-response curve, but alpha-adrenergic blockade sigwere subjected to alpha-adrenergic blockade with phen- nificantly attenuated the response (P < .Ol). oxybenzamine hydrochloride (Dibenzyline, Smith, The results of isoproterenol injections are shown in Kline, & French, Philadelphia) 1.5 mg kg-’ diluted in 10 Fig 4. This adrenergic amine is a pure vasodilator and ml kg-’ normal saline, and half of the animals were resulted in a linearly dependent increase in blood flow subjected to beta-adrenergic blockade with propranolol with increasing doses. At the highest dose, blood flow (Inderal, Ayerst Laboratories, New York) at a dose of increased 2.06 & .39 ml min-’ kg-‘. Only beta-adrener0.5 mg kg-’ iv. Forty-five minutes following alpha-adregic blockade affected the responses (P < .OOl). nergic blockade and 30 min following beta-adrenergic blockade, the intra-arterial injections and infusions were repeated. The completeness of alpha-adrenergic blockade was assessedby the ability of phenoxybenzamine to reverse the systemic pressor response to intraveFLOWMETER nous epinephrine (1.0 pg kg-‘), and the completeness of beta blockade was assessedby the ability of propranolol to significantly reduce the systemic hypotensive effects INFUSION CnTHETEd of intravenous isoproterenol (1.0 pg kg-‘). Because of the wide range in animal size, data are expressed as the mean 2 standard error for absolute flow per kilogram body weight for the infusion data and change in absolute flow per kilogram for the dose-response data. Statistical analysis for the dose-response data was performed using regression analysis for heterogeneity of two lines. The infusion data were analyzed for significance by a multivariant analysis (Hotelling’s T* test for multiple comparisons). FIG. 1. Experimental preparation. RESULTS
Control gastric blood flow was 3.7 2 .3 ml min-’ kg-‘; an absolute flow of 74 ml min-’ for mean body weight. This was not significantly changed by either alpha- or beta-adrenergic blockade. Neither control arterial pressure, 126 t 3 mmHg, nor portal pressure, 7 2 1 mmHg, was significantly altered following adrenergic blockade. The effectiveness of alpha-adrenergic blockade was assessedby intravenous injections of epinephrine. Prior to alpha blockade, intravenous epinephrine had an arterial pressor effect of 35 IT 5 mmHg and, following blockade, this dose of epinephrine caused a fall in arterial pressure of 39 ? 4 mmHg (P < .Ol). Intravenous isoproterenol was used to evaluate beta-adrenergic blockade. Prior to blockade arterial pressure fell 31 2 5 mmHg and after blockade only 4 ? 6 mmHg (P < .Ol). Since intra-arterial infusions of catecholamines at the dose employed minimally altered arterial pressure (less than 5 mmHg) or portal pressure (less than 2 mmHg), calcu-
DOSE,
pg
kg-’
CONTROL C+ -J PHENOXYBENZAMINE,I 5mgkg-! cs.4 PROPRANOLOL, 0 5 m g kg-‘, I V
I V
FIG. 2. Effects of intra-arterial injections of epinephrine on baboon gastric blood flow before (n = 10) and after alpha- (n = 5) or beta- (n = 5) adrenergic blockade. Ordinate: absolute change in flow in ml min-’ kg-’ (body wt).
Downloaded from www.physiology.org/journal/ajplegacy by ${individualUser.givenNames} ${individualUser.surname} (130.070.008.131) on January 11, 2019.
348
ZINNER, EPINEPHRINE,
KERR,
AND
O.O5pqkq-bin”
REYNOLDS ]
4.00 T 2 3.00 T .-: z 2.00 DOSE, pg kg-’ CONTROL o-aPHENOXYBENZAMINE, b-4 PROPRANOLOL,0.5 FIG. 3. Effects of intraarterial baboon gastric blood flow before beta- (n = 5) adrenergic blockade. in ml min-* kg-’ (body wt).
injections of norepinephrine on (n = 10) and after alpha- (n = 5) or Ordinate: absolute change in flow
L
a” 0
A/ d.M
10-2 DOSE,
e-a b--4
TT
L
0
T
0 0’f 0 0 .00 .) ./--P./-- P J./ -.-.-. 10-3
0
5
10-l
IO
TIME,
-CONTROL
2.00
.: 2
I .oo
1.5 m g kg-‘, I.V. I .V.
m g kg-‘,
I -i -2
6
e-0
PHENOXYBENZAMINE,
&-.A
PROPRANOLOL,
MINUTES 1.5 mg kg-l,
0.5 mq kg-l,
i.v.
i .v.
FIG. 5. Effects of intra-arterial infusion of epinephrine on baboon gastric blood flow before (n = 10) and after alpha- (n = 5) or beta- (n = 5) adrenergic blockade. Ordinate; ‘absolute flow in ml min-’ kg-’ (body wt).
IO0
NOREPINEPHRINE,
0.05 pg kg-’ m 1n-’
pq kq-’
CONTROL PHENOXYBENZAMINE, I.5 mq kq-‘, PROPRANOLOL, 0.5 mq kg-‘, I .V.
4.00
-
2 3.00 -I
-
I.V.
FIG. 4. Effects of intra-arterial injections of isoproterenol on baboon gastric blood flow before (n = 10) and after alpha- (n = 5) or beta- (n = 5)’ adrenergic blockade. Ordinate: absolute change in flow in ml min-’ kg-’ (body wt).
Zntra-arterial infusions. Epinephrine infusion (Fig. 5) caused a pure vasoconstrictor response. Within 3 min flow fell to 1.27 t .32 ml min-’ kg-’ and remained decreased throughout the infusion with no evidence of autoregulatory escape. Following the termination of the lo-min infusion, there was a gradual return toward control value, and at 12 min the flow was 3.86 t .7 ml min-’ kg-l, a value not significantly different than control (3.71 5 .3 ml min-’ kg-‘). Beta-adrenergic blockade did not significantly alter the response, but alpha-adrenergic blockade did attenuate the decrease in flow (P < . 01) The results of the infusion of norepinephrine on gastric blood flow are seen in Fig. 6. As with epinephrine, norepinephrine resulted in pure vasoconstriction with no autoregulatory escape and no postinfusion overshoot. At this dose norepinephrine did not appear to be as effective a vasoconstrictor as epinephrine. At 3 min of infusion, flow had decreased to 2.18 t .28 ml min-’ kg? The effects of adrenergic blockade on the response to norepinephrine are similar to those described for epinephrine. Figure 7 demonstrates the effects of isoproterenol infusion. There was a sustained increase in flow during the infusion. The response was significantly attenuated by beta-adrenergic blockade (P < .Ol) but not significantly influenced by alpha-adrenergic blockade.
T
E ;E 2.00
-
6 1.00
-
o-
FIG.
boon betamin-’
‘I
I
0
I
I
-
CONTROL
o-o
PHENOXYBENZAMINE,
b-4
PROPRANOLOL,
‘I
I
5
TIME,
I
I
“’
IO
MINUTES I .5 mq kg-‘,
i.v.
0.5 mq kg-‘, i .v.
6. Effects of intra-arterial infusion of norepinephrine gastric blood flow before (n = 10) and after alpha(n = 5) adrenergic blockade. Ordinate: is absolute kg-’ (body wt).
on ba(n = 5) or flow in ml
DISCUSSION
The results presented here indicate a species difference in the adrenergic mechanisms of the gastric circulation between the dog and the subhuman primate. In our previous canine studies (12, l3), these mechanisms were evaluated independently in the left and right gastric circulations. However, in the present study it was not possible to reproduce that experimental design because neither the right nor the left gastric arterv in the ba-
Downloaded from www.physiology.org/journal/ajplegacy by ${individualUser.givenNames} ${individualUser.surname} (130.070.008.131) on January 11, 2019.
PRIMATE
GASTRIC
349
CIRCULATION ,
ISOPROTERENOL,
0.05
pq kg-’
min-’
v
2.00
FIG.
boon betamin+
/c 1
V
1
0
1
1
N
CONTROL
0-0
PHENOXYBENZAMINE,
b-d
PROPRANOLOL,
1
’
1
5
TIME,
I
I
I
’
IO
I
I
MINUTE
I.5 mg kg-‘,
i.v.
0.5 mg kg-‘, i.v.
7. Effects of intra-arterial infusion of isoproterenol on bagastric blood flow before (n = 10) and after alpha(n = 5) or blockade. Ordinate: absolute flow in ml (n = 5) adrenergic kg-l (body wt).
boons was large enough for accurate, individual measurements of blood flow. In the dog, intra-arterial infusion of epinephrine into the left gastric artery caused predominantly a vasodilator response while, when infused into the right gastric artery, it caused vasoconstriction accompanied by autoregulatory escape. The infusion of norepinephrine into either gastric artery caused vasoconstriction with autoregulatory escape. Upon cessation of the infusion of either epinephrine or norepinephrine, there occurred a significant increase in gastric blood flow or an “overshoot” phenomenon. The intra-arterial infusion of isoproterenol caused a dilator response in both gastric arteries of this species, with the response in the left gastric artery being of greater magnitude than that in the right gastric artery. In contrast, the infusion of either epinephrine or norepinephrine into the baboon gastric circulation resulted in a sustained vasoconstrictor response. There is neither evidence of autoregulatory escape over the lo-min infusion nor an overshoot response upon cessation of the delivery of the adrenergic amines. These responses are neither potentiated by beta-adrenergic blockade nor reversed to vasodilation by alpha-adrenergic blockade as is the case in the canine right gastric circulation under the same experimental conditions. This observation would suggest thatthe beta-adrenergic receptor mechanism in the primate gastric circulation exerts less influence than in the canine vasculature. There are betaadrenergic receptors present in the baboon gastric circu-
to lation as documented by the ability of isoproterenol cause sustained vasod i.lation when it is infused into th .at circul ation. However, even here there exists a species difference in that the dilator response in the baboon amounted to an increase of flow to approximately 170% of control, whereas in the previous dog studies an identical dose of isoproterenol increased left gastric blood flow to a value in excess of 500% of control. Intra-arterial infusions of vasoconstrictor agents for the control of gastrointestinal bleeding has gained acceptance. A recent prospective controlled clinical trial of the use of intra-arterial vasopressin in the treatment of upper gastrointestinal hemorrhage (2) reported vasopressin to be more effective in controlling hemorrhage from nonvariceal lesions and from varices than conventional therapy. However, the use of intra-arterial vasopressin is not without complications. Even using low doses of intra-arterial vasopressin, Conn et al. (2) reported an increased incidence of hypertension and bradycardia or other arrhythmias in the vasopressin-treated group of patients with gastrointestinal bleeding compared with conventionally treated patients. In addition, because of its antidiuretic effects, hyponatremia and water retention have been reported by numerous clinical investigators (1, 6). Whereas, these complications of vasopressin therapy do not occur in most patients, there are certain patients that are more susceptible, e.g., patients with arteriosclerotic cardiovascular diseases or impaired renal function. It would be useful to have an alternative vasoconstrictor to control upper gastrointestinal bleeding that did not have these possible complications. Vasopressin is partially metabolized by the liver, but the rate of disappearance of exogenously administered vasopressin is to a great extent dependent on urinary excretion (5). Unlike vasopressin, epinephrine and norepinephrine are rapidly metabolized by the normal liver. In the present study the doses of catecholamines employed produced significant vasoconstriction without altering systemic arterial pressure or heart rate. In preliminary studies in baboons, we infused vasopressin intra-arterially to produce comparable decreases in gastric blood flow. There was a significant increase in arterial pressure and bradycardia (unpublished observations). It would appear that the naturally occurring catecholamines, epinephrine and norepinephrine, are effective vasoconstrictors in the primate gastric circulation and might be useful alternatives to vasopressin for the control of upper gastrointestinal bleeding from gastric sources in appropriate patients. In conducting the research described in this report, the investigators adhered to the “Guide for Laboratory Animal Facilities and Care,” as promulgated by the Committee on the Guide for Laboratory Animal Facilities and Care of the Institute of Laboratory Animal Resources, National Academy of Sciences-National Research Council. Received
for publication
10 September
1975.
REFERENCES 1. ATHANASOULIS, IMBEMBO,
hemorrhage: tract. New
C. A., S. BAUM, A. C. WALTMAN, E. J. RING, A. T. J. V. SALM. Control of acute gastric mucosal intra-arterial infusion of posterior pituitary exEngl. J. Med. 290: 597-603, 1974.
AND
2. CONN, H. O., G. R. RAMSBY, E. H. STORER, M. G. MUTCHNICK, P. H. JOSHI, M. M. PHILLIPS, G. A. COHEN, G. N. FIELDS, AND,D. PETROSKI. Intra-arterial vasopressin in the treatment of upper gastrointestinal hemorrhage: a prospective controlled clinical
Downloaded from www.physiology.org/journal/ajplegacy by ${individualUser.givenNames} ${individualUser.surname} (130.070.008.131) on January 11, 2019.
350 trial. Gastroenterology 68: 211-221, 1975. 3. CUMMING, J. D., A. L. HAIGH, E. H. L. HARRIS, AND M. E. NUTT. A study of gastric secretion and blood flow in the anesthetized dog. J. Physiol., London 168: 219-233, 1963. 4. DELANEY, J. P., AND E. GRIM. Experimentally induced variations in canine gastric blood flow and its distribution. Am. J. Physiol. 208: 353-358, 1965. 5. LAWSON, H. G. Metabolism of antidiuretic hormones. Am. J. Med. 42: 713-744, 1967. 6. MARUBBIO, A. T., R. P. LOMBARDO, AND P. R. HOLT. Control of variceal bleeding by superior mesenteric artery pitressin perfusions -complications and indications. Am. J. Digest. Diseases 18: 539-543, 1973. 7. NUSBAUM, M., S. BAUM, P. SAKIYALAK, AND W. S. BLAKEMORE. Pharmacologic control of portal hypertension. Surgery 62: 299310, 1967. 8. ROSCH, J., C. T. DOTTER, AND R. W. ROSE. Selective arterial
ZINNER,
9.
10. 11.
12.
13.
KERR,
AND
REYNOLDS
infusion of vasoconstrictors in acute gastrointestinal bleeding. Radiology 99: 27-36, 1971. THOMPSON, J. E., AND J. R. VANE. Gastric secretion induced by histamine and its relationship to the rate of blood flow. J. Physiol., London 121: 433-444, 1953. WALDER, D. N. Arteriovenous anastomoses of the human stomach. CZin. Sci. 11: 59-71, 1952. WHITE, R. I., D. P. HARRINGTON, G. NOVAK, F. J. MILLER, F. A. GIARGIANA, AND R. N. SHEFF. Pharmacologic control of hemorrhagic gastritis: clinical and experimental results. Radiology 111: 549-557, 1974. ZINNER, M. J., J. C. KERR, AND D. G. REYNOLDS. Adrenergic mechanisms in the canine circulation. Am. J. Physiol. 229: 977982, 1975. ZINNER, M. J., J. C. KERR, AND D. G. REYNOLDS. Hemodynamic effects of intra-arterial infusions of catecholamines on the canine gastric circulation. Surgery 78: 381-388, 1975.
Downloaded from www.physiology.org/journal/ajplegacy by ${individualUser.givenNames} ${individualUser.surname} (130.070.008.131) on January 11, 2019.
Primate gastric catecholamines MICHAEL Division
circulation: effects of and adrenergic blockade
J. ZINNER, of Surgery,
JOHN
Walter
C. KERR,
Reed Army
AND
Institute
ZINNER, MICHAEL J., JOHN C. KERR, AND DAVID G. REYNoms. Primate gastric circulation: effects of catecholamines and adrenergic blockade. Am. J. Physiol. 230(2): 346350. 1976. -The effects of intra-arterial injections and infusions of epinephrine, norepinephrine, and isoproterenol on gastric blood flow were studied in anesthetized baboons, Blood flow was measured electromagnetically before and after adrenergic blockade. The results for injected epinephrine and norepinephrine indicate these agents to be pure vasoconstrictors in the primate gastric circulation, and this response is attenuated by alpha-adrenergic blockade with phenoxybenzamine. Isoproterenol is a pure vasodilator, and its response is attenuated following beta-adrenergic blockade with proprano101. Intra-arterial infusions of epinephrine and norepinephrine C.05 pg kg-’ min-‘) resulted in sustained vasoconstriction with no evidence of autoregulatory escape and no postinfusion “overshoot.” This study suggests that epinephrine and norepinephrine might provide alternatives to vasopressin as a vasoconstrictor for the control of upper gastrointestinal bleeding.
epinephrine; norepinephrine; flow; autoregulatory escape
isoproterenol;
gastric
blood
andblockade on the splanchnic circulation have been studied by several methods and in several species. Previous investigators have reported that epinephrine and norepinephrine exert both vascoconstrictor and vasodilator effects on the gastric circulation. Whereas Walder (10) reported epinephrine to be a vasodilator in excised human stomachs, Thompson and Vane (9) found intravenous infusions of epinephrine, at low dose, to be a dilator of the gastric circulation and a high dose of epinephrine to cause vasoconstriction. Using the same intravenous dose range as Thompson and Vane, Cummings et al. (3) concluded that epinephrine was strictly a dilator and norepinephrine a constrictor in the canine gastric circulation. Delaney and Grim (4), using 42K clearance as a measure of blood flow in the dog, concurred with the observations that intravenous epinephrine increased blood flow and intravenous norepinephrine decreased blood flow to the stomach. In recent studies using intraarterial injections (12) and infusions (13) of catecholamines, we observed epinephrine to stimulate both alpha- and beta-adrenergic receptors in the canine gastric circulation. In this species epinephrine proved to be a dilator in the left gastric circulation. Norepinephrine had both alpha- and beta-adrenergic properties but unlike epinephrine was predominantly a vasoconstrictor. THE
EFFECTS
OF ADRENERGIC
STIMULATION
DAVID
G. REYNOLDS
of Research,
Washington,
D.C. 20012
The diversity of opinions concerning the action of these catecholamines in the gastric circulation is undoubtedly related to differences in species, routes and methods of drug delivery, blood flow monitoring techniques, and dose of drug employed. The action of this class of drugs is important in light of the clinical introduction of selective intra-arterial infusions of vasoconstrictor agents to control gastrointestinal bleeding (7). Catecholamines, either alone or in combination with beta-adrenergic blockade, have been used in all regional splanchnic circulations (8, 11). On the basis of our observations in dogs (13) that intra-arterial infusions of epinephrine into the gastric circulation result in either vasodilation or brief vasoconstriction followed by autoregulatory escape, we cautioned the clinical use of these catecholamines to control bleeding from gastric sources. The present study evaluates the effects of intra-arterial injections and infusions of epinephrine, norepinephrine, and isoproterenol into the gastric circulation of the subhuman primate before and after differential adrenergic blockade. MATERIALS
AND
METHODS
Ten adult Papio anubis baboons of either sex weighing 13.4-28.3 kg (mean 19.9 2 1.5 kg) were immobilized with phencyclidine hydrochloride (Sernylan, Bio-ceutic Laboratories, St. Joseph) (1.0 mg kg-l, im). The animals were then maintained with light pentobarbital anesthesia throughout the study. Through a midline laparotomy, the celiac artery was exposed and the transducer (In Vivo Metric Systems, Los Angeles) of a gated sinewave electromagnetic blood flowmeter (Biotronex Laboratories, Inc., Silver Spring, Md.) was placed on the artery near its origin. A hydraulic occluder was placed distally on the artery for zero flow determinations. The common hepatic artery was isolated and cannulated with a PE-90 polyethylene catheter (Clay-Adams, Inc., New York) for intra-arterial delivery of drugs. A splenectomy was performed with careful attention to the maintenance of the short gastric blood vessels (Fig. 1). Catheters were inserted into the abdominal aorta through a femoral artery (PE-190) and into the portal vein (PE-90) through a branch of the superior mesenteric vein and connected to pressure transducers (Statham Instruments, Inc., Oxnard, Calif.) for monitoring arterial and portal pressures. A femoral vein was cannulated (PE-190) for intravenous administration of drugs. Pressure transducers were calibrated prior to each experiment with a mercury manometer. Blood flow trans-
Downloaded from www.physiology.org/journal/ajplegacy by ${individualUser.givenNames} ${individualUser.surname} (130.070.008.131) on January 11, 2019.
PRIMATE
GASTRIC
347
CIRCULATION
ducers were calibrated with whole blood in vitro and lated gastric vascular resistance would be reflected by checked in vivo. All parameters were recorded simultachanges in gastric blood flow. neously on a multichannel recorder (Sanborn, WalIntra-arterial injections. The effects of intra-arterial tham). injections of epinephrine in the baboon gastric circulaThe drugs injected and infused were epinephrine hytion are seen in Fig. 2. At the lowest dose, 10e3pg kg-‘, drochloride (The Vitarine Co., New York), norepinephepinephrine elicited a decrease in flow of 0.35 t .12 ml rine bitartrate (Levophed, Winthrop Laboratories, New min-’ kg-‘. As the dose was increased, the constrictor York), and isoproterenol (Isuprel, Winthrop). Using sa- response increased to 2.90 ? .36 ml min-’ kg-’ at the line as a diluent, drugs were injected intra-arterially as highest dose. Beta-adrenergic blockade had no signifia O.l-ml bolus in doses ranging logarithmically from cant effect on the responses, but alpha-adrenergic blocklop3 to 10’ pg (base) kg-’ followed by a l.O-ml flush of ade significantly reduced the response (P < .OOl>.Intrasaline. Intra-arterial infusions of drugs were delivered arterial injections of norepinephrine over the same dose at a rate of 0.05 pg (base) kg-’ min-’ for 10 min. The range (Fig. 3) resulted in similar but somewhat smaller dose of drug employed was selected to give maximum responses. At the highest dose, 1.0 pg kg-‘, the decrease reproducible responses in total gastric blood flow within gastric blood flow was 2.42 t .39 ml min-’ kg-‘. Betaout altering arterial or portal pressures. adrenergic blockade did not significantly influence the Following the control responses, half the animals dose-response curve, but alpha-adrenergic blockade sigwere subjected to alpha-adrenergic blockade with phen- nificantly attenuated the response (P < .Ol). oxybenzamine hydrochloride (Dibenzyline, Smith, The results of isoproterenol injections are shown in Kline, & French, Philadelphia) 1.5 mg kg-’ diluted in 10 Fig 4. This adrenergic amine is a pure vasodilator and ml kg-’ normal saline, and half of the animals were resulted in a linearly dependent increase in blood flow subjected to beta-adrenergic blockade with propranolol with increasing doses. At the highest dose, blood flow (Inderal, Ayerst Laboratories, New York) at a dose of increased 2.06 & .39 ml min-’ kg-‘. Only beta-adrener0.5 mg kg-’ iv. Forty-five minutes following alpha-adregic blockade affected the responses (P < .OOl). nergic blockade and 30 min following beta-adrenergic blockade, the intra-arterial injections and infusions were repeated. The completeness of alpha-adrenergic blockade was assessedby the ability of phenoxybenzamine to reverse the systemic pressor response to intraveFLOWMETER nous epinephrine (1.0 pg kg-‘), and the completeness of beta blockade was assessedby the ability of propranolol to significantly reduce the systemic hypotensive effects INFUSION CnTHETEd of intravenous isoproterenol (1.0 pg kg-‘). Because of the wide range in animal size, data are expressed as the mean 2 standard error for absolute flow per kilogram body weight for the infusion data and change in absolute flow per kilogram for the dose-response data. Statistical analysis for the dose-response data was performed using regression analysis for heterogeneity of two lines. The infusion data were analyzed for significance by a multivariant analysis (Hotelling’s T* test for multiple comparisons). FIG. 1. Experimental preparation. RESULTS
Control gastric blood flow was 3.7 2 .3 ml min-’ kg-‘; an absolute flow of 74 ml min-’ for mean body weight. This was not significantly changed by either alpha- or beta-adrenergic blockade. Neither control arterial pressure, 126 t 3 mmHg, nor portal pressure, 7 2 1 mmHg, was significantly altered following adrenergic blockade. The effectiveness of alpha-adrenergic blockade was assessedby intravenous injections of epinephrine. Prior to alpha blockade, intravenous epinephrine had an arterial pressor effect of 35 IT 5 mmHg and, following blockade, this dose of epinephrine caused a fall in arterial pressure of 39 ? 4 mmHg (P < .Ol). Intravenous isoproterenol was used to evaluate beta-adrenergic blockade. Prior to blockade arterial pressure fell 31 2 5 mmHg and after blockade only 4 ? 6 mmHg (P < .Ol). Since intra-arterial infusions of catecholamines at the dose employed minimally altered arterial pressure (less than 5 mmHg) or portal pressure (less than 2 mmHg), calcu-
DOSE,
pg
kg-’
CONTROL C+ -J PHENOXYBENZAMINE,I 5mgkg-! cs.4 PROPRANOLOL, 0 5 m g kg-‘, I V
I V
FIG. 2. Effects of intra-arterial injections of epinephrine on baboon gastric blood flow before (n = 10) and after alpha- (n = 5) or beta- (n = 5) adrenergic blockade. Ordinate: absolute change in flow in ml min-’ kg-’ (body wt).
Downloaded from www.physiology.org/journal/ajplegacy by ${individualUser.givenNames} ${individualUser.surname} (130.070.008.131) on January 11, 2019.
348
ZINNER, EPINEPHRINE,
KERR,
AND
O.O5pqkq-bin”
REYNOLDS ]
4.00 T 2 3.00 T .-: z 2.00 DOSE, pg kg-’ CONTROL o-aPHENOXYBENZAMINE, b-4 PROPRANOLOL,0.5 FIG. 3. Effects of intraarterial baboon gastric blood flow before beta- (n = 5) adrenergic blockade. in ml min-* kg-’ (body wt).
injections of norepinephrine on (n = 10) and after alpha- (n = 5) or Ordinate: absolute change in flow
L
a” 0
A/ d.M
10-2 DOSE,
e-a b--4
TT
L
0
T
0 0’f 0 0 .00 .) ./--P./-- P J./ -.-.-. 10-3
0
5
10-l
IO
TIME,
-CONTROL
2.00
.: 2
I .oo
1.5 m g kg-‘, I.V. I .V.
m g kg-‘,
I -i -2
6
e-0
PHENOXYBENZAMINE,
&-.A
PROPRANOLOL,
MINUTES 1.5 mg kg-l,
0.5 mq kg-l,
i.v.
i .v.
FIG. 5. Effects of intra-arterial infusion of epinephrine on baboon gastric blood flow before (n = 10) and after alpha- (n = 5) or beta- (n = 5) adrenergic blockade. Ordinate; ‘absolute flow in ml min-’ kg-’ (body wt).
IO0
NOREPINEPHRINE,
0.05 pg kg-’ m 1n-’
pq kq-’
CONTROL PHENOXYBENZAMINE, I.5 mq kq-‘, PROPRANOLOL, 0.5 mq kg-‘, I .V.
4.00
-
2 3.00 -I
-
I.V.
FIG. 4. Effects of intra-arterial injections of isoproterenol on baboon gastric blood flow before (n = 10) and after alpha- (n = 5) or beta- (n = 5)’ adrenergic blockade. Ordinate: absolute change in flow in ml min-’ kg-’ (body wt).
Zntra-arterial infusions. Epinephrine infusion (Fig. 5) caused a pure vasoconstrictor response. Within 3 min flow fell to 1.27 t .32 ml min-’ kg-’ and remained decreased throughout the infusion with no evidence of autoregulatory escape. Following the termination of the lo-min infusion, there was a gradual return toward control value, and at 12 min the flow was 3.86 t .7 ml min-’ kg-l, a value not significantly different than control (3.71 5 .3 ml min-’ kg-‘). Beta-adrenergic blockade did not significantly alter the response, but alpha-adrenergic blockade did attenuate the decrease in flow (P < . 01) The results of the infusion of norepinephrine on gastric blood flow are seen in Fig. 6. As with epinephrine, norepinephrine resulted in pure vasoconstriction with no autoregulatory escape and no postinfusion overshoot. At this dose norepinephrine did not appear to be as effective a vasoconstrictor as epinephrine. At 3 min of infusion, flow had decreased to 2.18 t .28 ml min-’ kg? The effects of adrenergic blockade on the response to norepinephrine are similar to those described for epinephrine. Figure 7 demonstrates the effects of isoproterenol infusion. There was a sustained increase in flow during the infusion. The response was significantly attenuated by beta-adrenergic blockade (P < .Ol) but not significantly influenced by alpha-adrenergic blockade.
T
E ;E 2.00
-
6 1.00
-
o-
FIG.
boon betamin-’
‘I
I
0
I
I
-
CONTROL
o-o
PHENOXYBENZAMINE,
b-4
PROPRANOLOL,
‘I
I
5
TIME,
I
I
“’
IO
MINUTES I .5 mq kg-‘,
i.v.
0.5 mq kg-‘, i .v.
6. Effects of intra-arterial infusion of norepinephrine gastric blood flow before (n = 10) and after alpha(n = 5) adrenergic blockade. Ordinate: is absolute kg-’ (body wt).
on ba(n = 5) or flow in ml
DISCUSSION
The results presented here indicate a species difference in the adrenergic mechanisms of the gastric circulation between the dog and the subhuman primate. In our previous canine studies (12, l3), these mechanisms were evaluated independently in the left and right gastric circulations. However, in the present study it was not possible to reproduce that experimental design because neither the right nor the left gastric arterv in the ba-
Downloaded from www.physiology.org/journal/ajplegacy by ${individualUser.givenNames} ${individualUser.surname} (130.070.008.131) on January 11, 2019.
PRIMATE
GASTRIC
349
CIRCULATION ,
ISOPROTERENOL,
0.05
pq kg-’
min-’
v
2.00
FIG.
boon betamin+
/c 1
V
1
0
1
1
N
CONTROL
0-0
PHENOXYBENZAMINE,
b-d
PROPRANOLOL,
1
’
1
5
TIME,
I
I
I
’
IO
I
I
MINUTE
I.5 mg kg-‘,
i.v.
0.5 mg kg-‘, i.v.
7. Effects of intra-arterial infusion of isoproterenol on bagastric blood flow before (n = 10) and after alpha(n = 5) or blockade. Ordinate: absolute flow in ml (n = 5) adrenergic kg-l (body wt).
boons was large enough for accurate, individual measurements of blood flow. In the dog, intra-arterial infusion of epinephrine into the left gastric artery caused predominantly a vasodilator response while, when infused into the right gastric artery, it caused vasoconstriction accompanied by autoregulatory escape. The infusion of norepinephrine into either gastric artery caused vasoconstriction with autoregulatory escape. Upon cessation of the infusion of either epinephrine or norepinephrine, there occurred a significant increase in gastric blood flow or an “overshoot” phenomenon. The intra-arterial infusion of isoproterenol caused a dilator response in both gastric arteries of this species, with the response in the left gastric artery being of greater magnitude than that in the right gastric artery. In contrast, the infusion of either epinephrine or norepinephrine into the baboon gastric circulation resulted in a sustained vasoconstrictor response. There is neither evidence of autoregulatory escape over the lo-min infusion nor an overshoot response upon cessation of the delivery of the adrenergic amines. These responses are neither potentiated by beta-adrenergic blockade nor reversed to vasodilation by alpha-adrenergic blockade as is the case in the canine right gastric circulation under the same experimental conditions. This observation would suggest thatthe beta-adrenergic receptor mechanism in the primate gastric circulation exerts less influence than in the canine vasculature. There are betaadrenergic receptors present in the baboon gastric circu-
to lation as documented by the ability of isoproterenol cause sustained vasod i.lation when it is infused into th .at circul ation. However, even here there exists a species difference in that the dilator response in the baboon amounted to an increase of flow to approximately 170% of control, whereas in the previous dog studies an identical dose of isoproterenol increased left gastric blood flow to a value in excess of 500% of control. Intra-arterial infusions of vasoconstrictor agents for the control of gastrointestinal bleeding has gained acceptance. A recent prospective controlled clinical trial of the use of intra-arterial vasopressin in the treatment of upper gastrointestinal hemorrhage (2) reported vasopressin to be more effective in controlling hemorrhage from nonvariceal lesions and from varices than conventional therapy. However, the use of intra-arterial vasopressin is not without complications. Even using low doses of intra-arterial vasopressin, Conn et al. (2) reported an increased incidence of hypertension and bradycardia or other arrhythmias in the vasopressin-treated group of patients with gastrointestinal bleeding compared with conventionally treated patients. In addition, because of its antidiuretic effects, hyponatremia and water retention have been reported by numerous clinical investigators (1, 6). Whereas, these complications of vasopressin therapy do not occur in most patients, there are certain patients that are more susceptible, e.g., patients with arteriosclerotic cardiovascular diseases or impaired renal function. It would be useful to have an alternative vasoconstrictor to control upper gastrointestinal bleeding that did not have these possible complications. Vasopressin is partially metabolized by the liver, but the rate of disappearance of exogenously administered vasopressin is to a great extent dependent on urinary excretion (5). Unlike vasopressin, epinephrine and norepinephrine are rapidly metabolized by the normal liver. In the present study the doses of catecholamines employed produced significant vasoconstriction without altering systemic arterial pressure or heart rate. In preliminary studies in baboons, we infused vasopressin intra-arterially to produce comparable decreases in gastric blood flow. There was a significant increase in arterial pressure and bradycardia (unpublished observations). It would appear that the naturally occurring catecholamines, epinephrine and norepinephrine, are effective vasoconstrictors in the primate gastric circulation and might be useful alternatives to vasopressin for the control of upper gastrointestinal bleeding from gastric sources in appropriate patients. In conducting the research described in this report, the investigators adhered to the “Guide for Laboratory Animal Facilities and Care,” as promulgated by the Committee on the Guide for Laboratory Animal Facilities and Care of the Institute of Laboratory Animal Resources, National Academy of Sciences-National Research Council. Received
for publication
10 September
1975.
REFERENCES 1. ATHANASOULIS, IMBEMBO,
hemorrhage: tract. New
C. A., S. BAUM, A. C. WALTMAN, E. J. RING, A. T. J. V. SALM. Control of acute gastric mucosal intra-arterial infusion of posterior pituitary exEngl. J. Med. 290: 597-603, 1974.
AND
2. CONN, H. O., G. R. RAMSBY, E. H. STORER, M. G. MUTCHNICK, P. H. JOSHI, M. M. PHILLIPS, G. A. COHEN, G. N. FIELDS, AND,D. PETROSKI. Intra-arterial vasopressin in the treatment of upper gastrointestinal hemorrhage: a prospective controlled clinical
Downloaded from www.physiology.org/journal/ajplegacy by ${individualUser.givenNames} ${individualUser.surname} (130.070.008.131) on January 11, 2019.
350 trial. Gastroenterology 68: 211-221, 1975. 3. CUMMING, J. D., A. L. HAIGH, E. H. L. HARRIS, AND M. E. NUTT. A study of gastric secretion and blood flow in the anesthetized dog. J. Physiol., London 168: 219-233, 1963. 4. DELANEY, J. P., AND E. GRIM. Experimentally induced variations in canine gastric blood flow and its distribution. Am. J. Physiol. 208: 353-358, 1965. 5. LAWSON, H. G. Metabolism of antidiuretic hormones. Am. J. Med. 42: 713-744, 1967. 6. MARUBBIO, A. T., R. P. LOMBARDO, AND P. R. HOLT. Control of variceal bleeding by superior mesenteric artery pitressin perfusions -complications and indications. Am. J. Digest. Diseases 18: 539-543, 1973. 7. NUSBAUM, M., S. BAUM, P. SAKIYALAK, AND W. S. BLAKEMORE. Pharmacologic control of portal hypertension. Surgery 62: 299310, 1967. 8. ROSCH, J., C. T. DOTTER, AND R. W. ROSE. Selective arterial
ZINNER,
9.
10. 11.
12.
13.
KERR,
AND
REYNOLDS
infusion of vasoconstrictors in acute gastrointestinal bleeding. Radiology 99: 27-36, 1971. THOMPSON, J. E., AND J. R. VANE. Gastric secretion induced by histamine and its relationship to the rate of blood flow. J. Physiol., London 121: 433-444, 1953. WALDER, D. N. Arteriovenous anastomoses of the human stomach. CZin. Sci. 11: 59-71, 1952. WHITE, R. I., D. P. HARRINGTON, G. NOVAK, F. J. MILLER, F. A. GIARGIANA, AND R. N. SHEFF. Pharmacologic control of hemorrhagic gastritis: clinical and experimental results. Radiology 111: 549-557, 1974. ZINNER, M. J., J. C. KERR, AND D. G. REYNOLDS. Adrenergic mechanisms in the canine circulation. Am. J. Physiol. 229: 977982, 1975. ZINNER, M. J., J. C. KERR, AND D. G. REYNOLDS. Hemodynamic effects of intra-arterial infusions of catecholamines on the canine gastric circulation. Surgery 78: 381-388, 1975.
Downloaded from www.physiology.org/journal/ajplegacy by ${individualUser.givenNames} ${individualUser.surname} (130.070.008.131) on January 11, 2019.
