- Email: [email protected]
Angiotensin II inhibits the forearm vascular response to increased arterial pressure in humans
246
JACC Vol. 25, No. 1 January 1995:246-50
PHYSIOLOGY
Angiotensin II Inhibits the Forearm Vascular Response to Increased Arterial Pressure in Humans STEVEN
R. G O L D S M I T H ,
MD, FACC, GREGORY
J. H A S K I N G ,
MD
Minneapolis, Minnesota
Objectives. This study tested the hypothesis that angiotensin II may inhibit the forearm vascular resistance response to an increase in arterial pressure in normal humans. Background. Angiotensin II inhibits baroreflex-mediated reductions in heart rate and peripheral sympathetic activity during increases in arterial pressure in experimental animals. If present in humans, such effects could contribute to the pathophysiologic role of angiotensin II in hypertension and heart failure. Methods. Two investigations were performed. In the first, forearm vascular resistance responses were compared during equipressor infusions of angiotensin II and phenylephrine. In the second, heart rate, forearm vascular resistance and systemic venous norepinephrine spillover responses were compared during head-down tilt and head-down tilt plus phenylephrine with concomitant angiotensin II or vehicle infusions. Results. In the first study, forearm vascular resistance increased from 44 - 12 (mean - SD) to 54 -+ 13 U (p < 0.05) during angiotensin II but did not change during phenylephrine infusions (39 -+ 8.5 to 40 _+ 14 U) that increased mean arterial pressure comparably (88 -+ 9.8 to 103 ± 14 mm Hg during angiotensin II, p < 0.001; 91 -+ 7.6 to 104 -+ 9.2 mm Hg during phenylephrine, p < 0.001). In the second study, the decrease in heart rate and
forearm vascular resistance during the combination of head-down tilt and phenylephrine were both attenuated during concomitant angiotensin II compared with vehicle infusions: AHR/AMAP = -2.2 beats/rain per mm Hg during vehicle and -0.87 beats/min per mm Hg during angiotensin H (p = 0.07); AFVR/AMAP = -2.8 U/ram Hg during vehicle and -0.19 U/mm Hg during angiotensin II (p = 0.01), where AHR = change in heart rate; AMAP = change in mean arterial pressure; and AFVR = change in forearm vascular resistance. Norepinephrine spillover declined during vehicle infusions (612 -+ 367 to 418 +- 196 ng/min, p < 0.05) but not during angiotensin II infusions despite a greater increase in mean arterial pressure when the subpressor angiotensin II was combined with head-down tilt and phenylephrine (6.0 -+ 7.0 mm Hg during vehicle; 14 -+ 9.4 mm Hg during angiotensin II, p < 0.01). Conclusions. Both pressor and nonpressor infusions of angiotensin II immediately inhibit the forearm vascular response to mild baroreflex loading in normal humans. If present over the long term, such effects could contribute to inappropriate peripheral resistance in diseases such as hypertension and congestive heart failure. (J Am Coil Cardiol 1995;25:246-50)
In normal humans, an increase in blood pressure caused by angiotensin II is not associated with appropriate bradycardia (1,2). Although it has been suggested that angiotensin II may directly stimulate sympathetic activity, subpressor angiotensin II infusions given to normal humans do not increase plasma norepinephrine or norepinephrine spillover (3). Pressor infusions of angiotensin II also have no effect on norepinephrine spillover (2) and exert a minimal effect on muscle sympathetic nerve activity when the pressor effect is blocked with nitroprusside (4). Because angiotensin II probably has no direct positive chronotropic effects (5), the diminished bradycardia during blood pressure elevation induced by angiotensin II is most likely a result of inhibition of the baroreflex response to the increase in arterial pressure. An inhibitory effect of angiotensin
lI on the cardiac response to baroreflex activation has been directly confirmed in experimental animals and is probably a central effect (6). Experimental studies in rabbits have also shown that pressor infusions of angiotensin II inhibit baroreflex-evoked decreases in peripheral sympathetic nerve activity (6). A similar observation was reported for muscle sympathetic nerve activity in humans (4). No human data are available regarding the potentially clinically important correlate of peripheral sympathetic nerve inhibition, namely, an inhibition of normal vasodilating responses to increased arterial pressure or blood volume in the presence of angiotensin II. Such inhibition could have significant consequences in clinical states associated with high angiotensin II levels, such as some forms of hypertension and congestive heart failure. The specific hypothesis of the current studies was that angiotensin II would inhibit the forearm vascular response to changes in arterial pressure or central blood volume in humans. We report the results of two separate studies. First are the measurements of forearm blood flow and resistance obtained, but not reported, in our previous study (2), which focused on heart rate and norepinephrine
From the Hennepin County Medical Center and the University of Minnesota, Minneapolis, Minnesota. Manuscript received April 25, 1994; revised manuscript received July 25, 1994, accepted July 27, 1994. Address for corresnondence: Dr. Steven R. Goldsmith, Hennepin County Medical Center, Cardiology Division,701 Park Avenue, Minneapolis, Minnesota 55415. ©1995 by the American College of Cardiology
0735-1097/95/$9.50 0735-1097(94)0030-S
JACC Vol. 25, No. 1 January 1995:246~0
G O L D S M I T H AND HASKING ANGIOTENSIN II AND VASCULAR PRESSURE RESPONSE
spillover responses to pressor infusions of angiotensin II. Second is a new study in which the primary focus was on the forearm vascular response to subpressor infusion of angiotensin II. We believe that the results of these investigations, taken together, strongly support an inhibitory effect of angiotensin II on the forearm vascular response to baroreflex activation in humans.
247
Assays. Norepinephrine kinetics were assessed using the tracer method of Esler et al. (7). At steady state, endogenous norepinephrine and infused H-3 norepinephrine should be cleared at equivalent rates. If there is no rerelease of tracer in the body, then
Norepinephrine spilloverto plasma Plasma norepinephrine (pg/ml)
Methods First study. We previously reported (2) that in response to elevations in arterial pressure produced by angiotensin II and phenylephrine, heart rate declined only with phenylephrine. Changes in norepinephrine spillover were comparable, suggesting that angiotensin II did not directly stimulate sympathetic activity. Therefore, the lack of bradycardia was attributable to inhibition of the chronotropic response to baroreflex activation. In that study, we also obtained measurements of forearm blood flow and calculations of forearm vascular resistance. These results along with the concomitant measurements of mean arterial pressure are reported here. The control measurements were those obtained after the initial steadystate infusion of hydrogen-3 (H-3) norepinephrine in the supine position, with the intervention being 15 rain of infusion of angiotensin II (5 ng/kg body weight per min), vehicle or phenylephrine titrated to match the increases in arterial pressure achieved by angiotensin II relative to vehicle. Second study. Eight normal subjects (seven men, one woman; 20 to 41 years old, mean 27) participated in the new study. After placement of intravenous catheters, the subjects rested quietly for at least 30 min on a tilt bed in a comfortable 15° head-up position. At the end of this period, heart rate, arterial pressure (by automated cuff, Omega 1400, In Vivo Research Instruments) and forearm blood flow (by cuff plethysmography, Hokanson Instruments) were measured. Blood was obtained for plasma norepinephrine and for use as a blank in the assay for norepinephrine kinetics (five subjects). In these subjects, a bolus of tritiated norepinephrine was given (12/zCi) followed by a 0.8-/~Ci/min infusion for 30 min. At the end of 30 min of tritiated norepinephrine, all measurements were reassessed. This point was taken as baseline. An infusion of either angiotensin II (2 ng/kg per min) or vehicle was then initiated in a double-blind previously determined randomized sequence and maintained for 30 min. Again, all variables were reassessed. Subjects were then tilted into a 30° head-down position for 15 rain, after which a phenylephrine infusion was begun for a final 15 min in the head-down position to additionally stimulate high-pressure baroreceptors. All variables were assessed after the tilt protocol alone and during tilt plus phenylephrine. After the completion of a 15-rain increase in mean arterial pressure during the phenylephrine infusions, the infusions were discontinued, intravenous catheters removed and subjects discharged for the day. Each subject then returned at least 1 week later and repeated the protocol with the alternate infusion (angiotensin II or vehicle) not given on day 1.
Infusionrate of H-3 norepinephrine (liters/ram) • H-3 norepinephrine in infusate(dpm/liter) Plasma H-3 norepinephrine (dpm/liter)
'
where dpm = disintegrations per minute. Knowing the infusion rate for H-3 norepinephrine, the concentration of infused H-3 norepinephrine, the plasma concentration of H-3 norepinephrine (both in @m/liter) and the concentration of endogenous plasma norepinephrine, one can then calculate the spillover rate of endogenous norepinephrine as follows: Norepinephrine spillover (ng/min) = Plasmanorepinephrine.Clearance of H-3 norepinephrine. Statistics. In the first study, mean arterial pressure, forearm blood flow and forearm vascular resistance data during infusions of angiotensin II, vehicle and phenylephrine were analyzed by analysis of variance for repeated measures. A similar analysis was then performed on the changes in each variable between each control and the subsequent intervention. This analysis of variance indicated significant variation in the changes. The individual pairs of responses (angiotensin II vs. vehicle, angiotensin II vs. phenylephrine and phenylcphrine vs. vehicle) were then analyzed by paired t test with correction for multiple comparisons by the least significant difference method. In the second study, the responses of heart rate, forearm blood flow, forearm vascular resistance, plasma norepinephfine, norepinephrine clearance and norepinephrine spillover were analyzed on each study day by analysis of variance for repeated measures. The between-day changes in each variable (e.g., the change in forearm blood flow between control and head-down tilt during vehicle vs. that during subpressor angiotensin II) were then compared by paired t test. The ratio of changes in heart rate to changes in mean arterial pressure and the ratio of changes in forearm vascular resistance to changes in mean arterial pressure during the phenylephrine infusions were also compared by paired t test because blood pressure was higher after phenylephrine and angiotensin II compared with phenylephrine and vehicle.
Results First study. Table 1 contains the results of the arterial
pressure and forearm blood flow measurements as well as the forearm vascular resistance calculations obtained but not reported in our earlier investigation. Figure 1 displays the
248
GOLDSMITH AND HASKING ANGIOTENSINII AND VASCULARPRESSURE RESPONSE
JACC Wol.25, No. l January 1995:246-50
Table 1. Responses to Infusions of Angiotensin II, Phenylephrine or Vehicle in Six Normal Volunteers in Supine Position Mean arterial pressure (mm Hg) Vehicle Angiotensin II Phenylephrine Forearm blood flow (ml/100 per min) Vehicle Angiotensin II Phenylephrine Forearm vascularresistance (U) Vehicle Angiotensin II Phenylephrine
Control
Infusion
85 _+12 88 ± 9.8 91 _+7.6
89 +_11 103 ± 14*t 104 _+9.2't
2.6 + 0.53 2.1 + 0.42 2.3 + 0.48
2.4 = 0.41 1.9 ± 0.29:~§ 2.9 ± 1.3
34 ± 10 44 _+12 39 +_8.5
38 ± 10:~ 54 _+13:~§ 40 _+14
is
AMAP lO mm/Hg
0 Vehicle
Phenyl
A-II
Vehlcle
Phenyl
A-II
20
*p < 0.001, :~p < 0.05 versus control, tp < 0.05 versus vehicle. §p < 0.05 versus vehicle and phenylephrine.Data presented are mean value _+SD.
15 changes in arterial pressure and forearm vascular resistance from control values during vehicle, angiotensin II and phenylephrine. Angiotensin II and phenylephrine increased mean arterial pressure comparably (angiotensin II 88 2 9.8 to 103 + 14 mm Hg, p < 0.001; phenylephrine 91 2 7.6 to 104 + 9.2 mm Hg, p < 0.001). Pressure did not change with vehicle. Forearm blood flow did not change with either vehicle or phenylephrine and declined slightly but significantly during angiotensin II (2.1 _+ 0.42 to 1.9 2 2.9 ml/100 ml per min, p < 0.05). Forearm vascular resistance increased slightly during vehicle, did not change with phenylephrine and rose markedly during angiotensin II (44 2 12 to 54 2 13 U, p < 0.05). Second study. Table 2 contains the data from the second study. Before head-down tilt, neither angiotensin II nor vehicle had any effect on any of the variables assessed. Head-down tilt alone failed to decrease heart rate, forearm vascular resistance or norepinephrine spillover on either study day. When phenylephrine was added to head-down tilt, mean arterial pressure increased during both angiotensin II and vehicle infusions. The change was greater with angiotensin II infusion (14 _+ 9.4 vs. 6.0 2 7.0 mm Hg, p < 0.05). Heart rate declined on both days, but as a function of the change in pressure, heart rate declined less with angiotensin II (-2.2 beats/ram Hg during vehicle, -0.87 beats/ram Hg during angiotensin II, p < 0.07) (Fig. 2). Forearm blood flow increased relative to control with both angiotensin II and vehicle. Forearm vascular resistance decreased appropriately with vehicle (48 2 15 to 40 2 11 U, p < 0.05) but for the group tended to increase with angiotensin II (41 2 17 to 43 2 21 U). As with heart rate, the change in forearm vascular resistance relative to mean arterial pressure was less during angiotensin II than during vehicle (-2.8 U/ram Hg during vehicle, -0.19 U/ram Hg during angiotensin II, p < 0.01) (Fig. 3). Norepinephrine spillover declined significantly when phenylephrine was combined with head-down tilt during vehicle infusion (612 2 367 to 418 2 396 ng/min, p < 0.05), but with angiotensin II it did not decrease despite the larger increase in mean arterial pressure.
A FVR
lO
0
Figure 1. Top, Response of mean arterial pressure (MAP) during vehicle, phenylephrine (Phenyl) and angiotensin II (A-II) infusions from the first study. Changes (A) in pressure were comparable during phenylephrine and angiotensin II infusions and significantly greater than during vehicle infusion. Bottom, Response of forearm vascular resistance (FVR) during vehicle, phenylephrine (Phenyl) and angiotensin II (A-II) infusions from the first study. Only the increase during angiotensin II infusion was significant. The response during vehicle and phenylephrine infusions was not different. A = change.
Discussion Present studies. The results of these studies are consistent with the hypothesis that modestly pressor (first study) and nonpressor (second study) infusions of angiotensin II inhibit peripheral vascular responses to activation of baroreflexes in normal humans. In the first study, despite comparable elevation of mean arterial pressure by angiotensin II and phenylephrine, forearm vascular resistance did not decrease during angiotensin II hut actually increased. During systemic infusion of any pressor, forearm vascular responses would be the result of the balance between direct vasoconstriction and withdrawal of endogenous vascular tone as arterial pressure increases and baroreflex activation occurs. Because there was no change in forearm vascular resistance during phenylephrine, these effects presumably balanced. With angiotensin II, forearm vascular resistance increased markedly during the same change in mean
JACC Vol. 25, No. l January 1995:246-50
G O L D S M I T H AND HASKING ANGIOTENSIN It AND V A S C U L A R PRESSURE RESPONSE
249
Table 2. Responses to Vehicle and Angiotensin II Alone and During Head-Down Tilt and Head-Down Tilt With Phenylcphrine Infusion and Comparison With Vehicle in Eight Normal Subjects
Heart rate (beats/min) Vehicle Angiotensin II Mean arterial pressure (mm Hg) Vehicle Angiotensin II Forearm blood flow (ml/100 ml per min) Vehicle Angiotensin II Forearm vascular resistance (U) Vehicle Angiotensin I1 Norepinephrine spillover (ng/min) Vehicle Angiotensin II
Control
Infusion
HeadDown Tilt
Phenylephrineand Head-DownTilt
61 +_ 11 60 -+ 6.9
60 _+11 58 _+9.0
58 _+9.5 59 _+15
54 +_11" 51 _+7.6*
85 _+7.8 88 -+ 6.7
87 -+ 8.1 93 _+8.2
87 -+ 7.6 93 _+9.3
91 +_6.9* 102 ÷ 11'
1.8 _+0.27 2.4 _+0.78
2.1 _+0.62 2.4 _+0.89
2.1 _+0.58 3.0 _+1.3'
2.4 - 0.67¢ 2.7 -+ 0.99
48 _+9.3 41 -- 17
44 - 15 44 _+22
44 _ 12 37 _+ 18
40 _+l l t 43 _+21
612 ÷ 367 395 -- 105
637 +_387 451 _+283
548 +_265 518 _+445
418 + 196t 476 _+316
*p < 0.001, ?p < 0.05 versus control. Data presented are mean value _+SD.
arterial pressure. This result would be consistent with inhibition of baroreflex-mediated suppression of endogenous vasoconstrictor tone. In the second study, head-down tilt failed to produce a decrease in forearm vascular resistance or norepinephrine spillover on the vehicle day. This stimulus is predominantly to the cardiac reflexes because it increases central venous volume and pressure but not arterial pressure (8). When head-down tilt was combined with phenylephrine, arterial pressure increased, and heart rate, forearm vascular resistance and norepinephrine spillover all declined significantly, as expected during vehicle infusions. Heart rate also declined during angiotensin II, but the change in mean arterial pressure was greater when phenylephrine was added to the subpressor
Figure 2. Ratio of changes (A) in heart rate (HR) and mean arterial pressure (MAP) during head-down flit combined with phenylephrine infusion in the second study. As a function of change in mean arterial pressure, the decrease in heart rate was greater when the loading stimulus was applied during vehicle infusion than during angiotensin II (A-II) infusion.
angiotensin II than it was with vehicle. The change in heart rate relative to arterial pressure, although of borderline statistical significance between angiotensin II and vehicle days (Fig. 2), was less during angiotensin II, suggesting that as in previous studies (2), angiotensin II attenuated baroreflex-mediated responses to the increase in arterial pressure. Forearm vascular resistance declined appropriately when phenylephrine was given in the presence of the vehicle for angiotensin II but failed to decrease during phenylephrine plus angiotensin II despite the higher mean arterial pressure. In this case, the change in resistance relative to pressure was markedly less with angiotensin II than with vehicle (Fig. 3). The response of norepinephrine spillover (Table 2) paralleled that of forearm vascular
Figure 3. Ratio of the changes (A) in forearm vascular resistance (FVR) and mean arterial pressure (MAP) during head-down tilt combined with phenylephrine infusion in the second study. As a function of change in mean arterial pressure, the decrease in forearm vascular resistance was greater when the loading stimulus was applied during vehicle infusion than during angiotensin II (A-II) infusion.
-1 -1
-2 AFVRIAMAP
AHR/AMAP -2
-3 ,.J
-3 *p<.07 -4
-5
*p<.OS
-6 Vehicle
A-II
Vehicle
A-II
250
GOLDSMITH AND HASKING ANGIOTENSIN II AND VASCULAR PRESSURE RESPONSE
resistance, with a significant decline during head-down tilt and phenylephrine in the presence of vehicle and no change in the presence of angiotensin II despite a higher pressure. It is therefore likely that in these studies, angiotensin II attenuated the heart rate and the peripheral vascular and systemic sympathetic aspects of the efferent limb of the baroreflex. We should note that we have not performed a complete test of baroreflex gain and set point because we have only one interventional time point, but because pressure began at similar levels before phenylephrine on each study day, it is unlikely that we began at markedly different positions on the operating curve for the baroreceptors, making our comparison of the ratios of heart rate and forearm resistance to pressure a reasonable one. A more complete study over a wide range of pressure would be required to determine whether the effect of angiotensin II was on the gain or threshold of baroreflex function. Previous studies. These results are consistent with an observation by Matsukawa et al. (4), who reported attenuation of the muscle sympathetic nerve response to increased arterial pressure induced by angiotensin II relative to phenylephrine. Those investigators found a profound difference in muscle sympathetic nerve activity suppression after equipressor infusions of angiotensin II and phenylephrine yet nearly no effect of angiotensin II on muscle sympathetic nerve activity when blood pressure was held constant with nitroprusside. The latter result is consistent with our previous and current observations regarding the absence of effects of subpressor angiotensin II infusions on systemic norepinephrine spillover. There is, therefore, little chance that a direct stimulation of sympathetic activity by angiotensin II at these infusion rates confounded our interpretation of the changes during baroreflex activation. Our data combined with those of Matsukawa et al. suggest an inhibitory effect of angiotensin II on heart rate and systemic sympathetic and peripheral vascular responses to baroreceptor loading in normal humans. The effect on heart rate and sympathetic responses seems clear, but it remains possible that at least part of the attenuation of the forearm vascular responses could have been a result of amplification by angiotensin II of the effects of the phenylephrine in the forearm. However, two arguments may be made against this possibility. First, if such an effect occurred and contributed to the higher arterial pressure with an intact baroreflex, systemic norepinephrine spillover should have declined. Yet norepinephrine spillover declined when pressure increased with phenylephrine plus vehicle but did not decrease when angiotensin II was added to phenylephrine despite the higher pressure (Table 2). Second, in a previous study using 2 ng/kg per min infusions of angiotensin II (3), forearm vascular resistance responses were not potentiated during upright tilt in the presence of angiotensin II compared with vehicle. In that study, during upright tilt with vehicle infusions, forearm vascular resistance increased from 39 _+ 10 to 52 _+ 17 U (p < 0.05), whereas in the presence of 2 ng/kg per min of angiotensin II, it increased from 58 _+ 16 to 68 _+ 23 U (p < 0.05). The increase in forearm vascular resistance during upright tilt was no different between days.
JACC gol. 25, No. 1 January 1995:246-50
There is no obvious reason why angiotensin II would amplify an exogenous alpha-adrenergic stimulus (phenylephrine) in the forearm circulation during head-down tilt but not amplify an endogenous alpha-adrenergic stimulus in the forearm during the relatively potent sympathetic stimulation offered by upright tilt. The most likely explanation for the difference in forearm vascular responses is, therefore, baroreflex inhibition if neither direct sympathostimulation nor interaction with phenylephrine is probable. The results of these studies are in complete accord with several studies suggesting that angiotensin II inhibits baroreflex responses to baroreceptor loading in experimental animals (6). The most relevant single study is probably by Guo and Abboud (9), who documented that angiotensin II inhibited cardiac and peripheral sympathetic responses to an increase in pressure but did not restrain these responses during hypotension. Because, as noted, we also found no effect of either subpressor or pressor angiotensin II infusions on heart rate or systemic sympathetic or forearm vascular responses to baroreflex unloading with head-up tilt, it seems that angiotensin II interacts with the baroreflex only when pressure increases. Not resolved in any species is whether angiotensin II can inhibit the response to normotensive loading of the cardiopulmonary baroreceptors. Summary. Previous and present data do not support a direct effect of systemic infusions of angiotensin II on sympathetic activity in normal humans. Systemic infusion of angiotensin II is also without apparent direct effect on systemic norepinephrine spillover in congestive heart failure (10). However, baroreflex inhibitory effects of angiotensin II could contribute significantly to sympathetic dysregulation in hypertension and heart failure, representing an indirect mechanism by which angiotensin II contributes to volume expansion and excess vasoconstriction in these diseases.
References 1. Mace PJE, Watson RDS, Skan W, Littler WA. Inhibition of the baroreceptor heart rate reflex by angiotensin II in normal man. Cardiovasc Res 1985;19: 525-7. 2. Goldsmith SR, Hasking GJ. Effect of a pressor infusion of angiotensin II on sympathetic activity and heart rate in normal humans. Circ Res 1991;68: 263-8. 3. Goldsmith SR, Hasking GJ. Subpressor infusions of angiotensin II do not stimulate sympathetic activity in humans. Am J Physiol 1990;258:H179-82. 4. Matsukawa T, Gotoh E, Minimijawa K, et aL Effects of intravenous infusions of angiotensin II on muscle sympathetic nerve activity in humans. Am J Physiol 1991;261:690-6. 5. James TN. Absence of direct chronotropic action of angiotensin infused into the sinus node artery. Am J Physiol 1965;209:571-6. 6. Reid 1. Interactions between angiotensin II, sympathetic nervous system, and baroreceptor reflexes in regulation of blood pressure. Am J Physiol 1992; 262:E763 78. 7. Esler M, Jackman G, Bobik A, et al. Determination of norepinephrine apparent spillover rate and clearance in humans. Life Sci 1979;25:1461-70. 8. Goldsmith SR, Cowley AW, Francis GS, Cohn JN. Effects of increased central venous and aortic pressure on plasma vasopressin in man. Am J Physiol 1984;246:H647-51. 9. Guo GB, Abboud FM. Angiotensin II attenuates baroreflex control of heart rate and sympathetic activity. Am J Physiol 1984;246:H80-9. 10. Goldsmith SR, Hasking GJ. Angiotensin-II and sympathetic nervous activity in congestive heart failure. J Am Coil Cardiol 1993;21:1107-13.
JACC Vol. 25, No. 1 January 1995:246-50
PHYSIOLOGY
Angiotensin II Inhibits the Forearm Vascular Response to Increased Arterial Pressure in Humans STEVEN
R. G O L D S M I T H ,
MD, FACC, GREGORY
J. H A S K I N G ,
MD
Minneapolis, Minnesota
Objectives. This study tested the hypothesis that angiotensin II may inhibit the forearm vascular resistance response to an increase in arterial pressure in normal humans. Background. Angiotensin II inhibits baroreflex-mediated reductions in heart rate and peripheral sympathetic activity during increases in arterial pressure in experimental animals. If present in humans, such effects could contribute to the pathophysiologic role of angiotensin II in hypertension and heart failure. Methods. Two investigations were performed. In the first, forearm vascular resistance responses were compared during equipressor infusions of angiotensin II and phenylephrine. In the second, heart rate, forearm vascular resistance and systemic venous norepinephrine spillover responses were compared during head-down tilt and head-down tilt plus phenylephrine with concomitant angiotensin II or vehicle infusions. Results. In the first study, forearm vascular resistance increased from 44 - 12 (mean - SD) to 54 -+ 13 U (p < 0.05) during angiotensin II but did not change during phenylephrine infusions (39 -+ 8.5 to 40 _+ 14 U) that increased mean arterial pressure comparably (88 -+ 9.8 to 103 ± 14 mm Hg during angiotensin II, p < 0.001; 91 -+ 7.6 to 104 -+ 9.2 mm Hg during phenylephrine, p < 0.001). In the second study, the decrease in heart rate and
forearm vascular resistance during the combination of head-down tilt and phenylephrine were both attenuated during concomitant angiotensin II compared with vehicle infusions: AHR/AMAP = -2.2 beats/rain per mm Hg during vehicle and -0.87 beats/min per mm Hg during angiotensin H (p = 0.07); AFVR/AMAP = -2.8 U/ram Hg during vehicle and -0.19 U/mm Hg during angiotensin II (p = 0.01), where AHR = change in heart rate; AMAP = change in mean arterial pressure; and AFVR = change in forearm vascular resistance. Norepinephrine spillover declined during vehicle infusions (612 -+ 367 to 418 +- 196 ng/min, p < 0.05) but not during angiotensin II infusions despite a greater increase in mean arterial pressure when the subpressor angiotensin II was combined with head-down tilt and phenylephrine (6.0 -+ 7.0 mm Hg during vehicle; 14 -+ 9.4 mm Hg during angiotensin II, p < 0.01). Conclusions. Both pressor and nonpressor infusions of angiotensin II immediately inhibit the forearm vascular response to mild baroreflex loading in normal humans. If present over the long term, such effects could contribute to inappropriate peripheral resistance in diseases such as hypertension and congestive heart failure. (J Am Coil Cardiol 1995;25:246-50)
In normal humans, an increase in blood pressure caused by angiotensin II is not associated with appropriate bradycardia (1,2). Although it has been suggested that angiotensin II may directly stimulate sympathetic activity, subpressor angiotensin II infusions given to normal humans do not increase plasma norepinephrine or norepinephrine spillover (3). Pressor infusions of angiotensin II also have no effect on norepinephrine spillover (2) and exert a minimal effect on muscle sympathetic nerve activity when the pressor effect is blocked with nitroprusside (4). Because angiotensin II probably has no direct positive chronotropic effects (5), the diminished bradycardia during blood pressure elevation induced by angiotensin II is most likely a result of inhibition of the baroreflex response to the increase in arterial pressure. An inhibitory effect of angiotensin
lI on the cardiac response to baroreflex activation has been directly confirmed in experimental animals and is probably a central effect (6). Experimental studies in rabbits have also shown that pressor infusions of angiotensin II inhibit baroreflex-evoked decreases in peripheral sympathetic nerve activity (6). A similar observation was reported for muscle sympathetic nerve activity in humans (4). No human data are available regarding the potentially clinically important correlate of peripheral sympathetic nerve inhibition, namely, an inhibition of normal vasodilating responses to increased arterial pressure or blood volume in the presence of angiotensin II. Such inhibition could have significant consequences in clinical states associated with high angiotensin II levels, such as some forms of hypertension and congestive heart failure. The specific hypothesis of the current studies was that angiotensin II would inhibit the forearm vascular response to changes in arterial pressure or central blood volume in humans. We report the results of two separate studies. First are the measurements of forearm blood flow and resistance obtained, but not reported, in our previous study (2), which focused on heart rate and norepinephrine
From the Hennepin County Medical Center and the University of Minnesota, Minneapolis, Minnesota. Manuscript received April 25, 1994; revised manuscript received July 25, 1994, accepted July 27, 1994. Address for corresnondence: Dr. Steven R. Goldsmith, Hennepin County Medical Center, Cardiology Division,701 Park Avenue, Minneapolis, Minnesota 55415. ©1995 by the American College of Cardiology
0735-1097/95/$9.50 0735-1097(94)0030-S
JACC Vol. 25, No. 1 January 1995:246~0
G O L D S M I T H AND HASKING ANGIOTENSIN II AND VASCULAR PRESSURE RESPONSE
spillover responses to pressor infusions of angiotensin II. Second is a new study in which the primary focus was on the forearm vascular response to subpressor infusion of angiotensin II. We believe that the results of these investigations, taken together, strongly support an inhibitory effect of angiotensin II on the forearm vascular response to baroreflex activation in humans.
247
Assays. Norepinephrine kinetics were assessed using the tracer method of Esler et al. (7). At steady state, endogenous norepinephrine and infused H-3 norepinephrine should be cleared at equivalent rates. If there is no rerelease of tracer in the body, then
Norepinephrine spilloverto plasma Plasma norepinephrine (pg/ml)
Methods First study. We previously reported (2) that in response to elevations in arterial pressure produced by angiotensin II and phenylephrine, heart rate declined only with phenylephrine. Changes in norepinephrine spillover were comparable, suggesting that angiotensin II did not directly stimulate sympathetic activity. Therefore, the lack of bradycardia was attributable to inhibition of the chronotropic response to baroreflex activation. In that study, we also obtained measurements of forearm blood flow and calculations of forearm vascular resistance. These results along with the concomitant measurements of mean arterial pressure are reported here. The control measurements were those obtained after the initial steadystate infusion of hydrogen-3 (H-3) norepinephrine in the supine position, with the intervention being 15 rain of infusion of angiotensin II (5 ng/kg body weight per min), vehicle or phenylephrine titrated to match the increases in arterial pressure achieved by angiotensin II relative to vehicle. Second study. Eight normal subjects (seven men, one woman; 20 to 41 years old, mean 27) participated in the new study. After placement of intravenous catheters, the subjects rested quietly for at least 30 min on a tilt bed in a comfortable 15° head-up position. At the end of this period, heart rate, arterial pressure (by automated cuff, Omega 1400, In Vivo Research Instruments) and forearm blood flow (by cuff plethysmography, Hokanson Instruments) were measured. Blood was obtained for plasma norepinephrine and for use as a blank in the assay for norepinephrine kinetics (five subjects). In these subjects, a bolus of tritiated norepinephrine was given (12/zCi) followed by a 0.8-/~Ci/min infusion for 30 min. At the end of 30 min of tritiated norepinephrine, all measurements were reassessed. This point was taken as baseline. An infusion of either angiotensin II (2 ng/kg per min) or vehicle was then initiated in a double-blind previously determined randomized sequence and maintained for 30 min. Again, all variables were reassessed. Subjects were then tilted into a 30° head-down position for 15 rain, after which a phenylephrine infusion was begun for a final 15 min in the head-down position to additionally stimulate high-pressure baroreceptors. All variables were assessed after the tilt protocol alone and during tilt plus phenylephrine. After the completion of a 15-rain increase in mean arterial pressure during the phenylephrine infusions, the infusions were discontinued, intravenous catheters removed and subjects discharged for the day. Each subject then returned at least 1 week later and repeated the protocol with the alternate infusion (angiotensin II or vehicle) not given on day 1.
Infusionrate of H-3 norepinephrine (liters/ram) • H-3 norepinephrine in infusate(dpm/liter) Plasma H-3 norepinephrine (dpm/liter)
'
where dpm = disintegrations per minute. Knowing the infusion rate for H-3 norepinephrine, the concentration of infused H-3 norepinephrine, the plasma concentration of H-3 norepinephrine (both in @m/liter) and the concentration of endogenous plasma norepinephrine, one can then calculate the spillover rate of endogenous norepinephrine as follows: Norepinephrine spillover (ng/min) = Plasmanorepinephrine.Clearance of H-3 norepinephrine. Statistics. In the first study, mean arterial pressure, forearm blood flow and forearm vascular resistance data during infusions of angiotensin II, vehicle and phenylephrine were analyzed by analysis of variance for repeated measures. A similar analysis was then performed on the changes in each variable between each control and the subsequent intervention. This analysis of variance indicated significant variation in the changes. The individual pairs of responses (angiotensin II vs. vehicle, angiotensin II vs. phenylephrine and phenylcphrine vs. vehicle) were then analyzed by paired t test with correction for multiple comparisons by the least significant difference method. In the second study, the responses of heart rate, forearm blood flow, forearm vascular resistance, plasma norepinephfine, norepinephrine clearance and norepinephrine spillover were analyzed on each study day by analysis of variance for repeated measures. The between-day changes in each variable (e.g., the change in forearm blood flow between control and head-down tilt during vehicle vs. that during subpressor angiotensin II) were then compared by paired t test. The ratio of changes in heart rate to changes in mean arterial pressure and the ratio of changes in forearm vascular resistance to changes in mean arterial pressure during the phenylephrine infusions were also compared by paired t test because blood pressure was higher after phenylephrine and angiotensin II compared with phenylephrine and vehicle.
Results First study. Table 1 contains the results of the arterial
pressure and forearm blood flow measurements as well as the forearm vascular resistance calculations obtained but not reported in our earlier investigation. Figure 1 displays the
248
GOLDSMITH AND HASKING ANGIOTENSINII AND VASCULARPRESSURE RESPONSE
JACC Wol.25, No. l January 1995:246-50
Table 1. Responses to Infusions of Angiotensin II, Phenylephrine or Vehicle in Six Normal Volunteers in Supine Position Mean arterial pressure (mm Hg) Vehicle Angiotensin II Phenylephrine Forearm blood flow (ml/100 per min) Vehicle Angiotensin II Phenylephrine Forearm vascularresistance (U) Vehicle Angiotensin II Phenylephrine
Control
Infusion
85 _+12 88 ± 9.8 91 _+7.6
89 +_11 103 ± 14*t 104 _+9.2't
2.6 + 0.53 2.1 + 0.42 2.3 + 0.48
2.4 = 0.41 1.9 ± 0.29:~§ 2.9 ± 1.3
34 ± 10 44 _+12 39 +_8.5
38 ± 10:~ 54 _+13:~§ 40 _+14
is
AMAP lO mm/Hg
0 Vehicle
Phenyl
A-II
Vehlcle
Phenyl
A-II
20
*p < 0.001, :~p < 0.05 versus control, tp < 0.05 versus vehicle. §p < 0.05 versus vehicle and phenylephrine.Data presented are mean value _+SD.
15 changes in arterial pressure and forearm vascular resistance from control values during vehicle, angiotensin II and phenylephrine. Angiotensin II and phenylephrine increased mean arterial pressure comparably (angiotensin II 88 2 9.8 to 103 + 14 mm Hg, p < 0.001; phenylephrine 91 2 7.6 to 104 + 9.2 mm Hg, p < 0.001). Pressure did not change with vehicle. Forearm blood flow did not change with either vehicle or phenylephrine and declined slightly but significantly during angiotensin II (2.1 _+ 0.42 to 1.9 2 2.9 ml/100 ml per min, p < 0.05). Forearm vascular resistance increased slightly during vehicle, did not change with phenylephrine and rose markedly during angiotensin II (44 2 12 to 54 2 13 U, p < 0.05). Second study. Table 2 contains the data from the second study. Before head-down tilt, neither angiotensin II nor vehicle had any effect on any of the variables assessed. Head-down tilt alone failed to decrease heart rate, forearm vascular resistance or norepinephrine spillover on either study day. When phenylephrine was added to head-down tilt, mean arterial pressure increased during both angiotensin II and vehicle infusions. The change was greater with angiotensin II infusion (14 _+ 9.4 vs. 6.0 2 7.0 mm Hg, p < 0.05). Heart rate declined on both days, but as a function of the change in pressure, heart rate declined less with angiotensin II (-2.2 beats/ram Hg during vehicle, -0.87 beats/ram Hg during angiotensin II, p < 0.07) (Fig. 2). Forearm blood flow increased relative to control with both angiotensin II and vehicle. Forearm vascular resistance decreased appropriately with vehicle (48 2 15 to 40 2 11 U, p < 0.05) but for the group tended to increase with angiotensin II (41 2 17 to 43 2 21 U). As with heart rate, the change in forearm vascular resistance relative to mean arterial pressure was less during angiotensin II than during vehicle (-2.8 U/ram Hg during vehicle, -0.19 U/ram Hg during angiotensin II, p < 0.01) (Fig. 3). Norepinephrine spillover declined significantly when phenylephrine was combined with head-down tilt during vehicle infusion (612 2 367 to 418 2 396 ng/min, p < 0.05), but with angiotensin II it did not decrease despite the larger increase in mean arterial pressure.
A FVR
lO
0
Figure 1. Top, Response of mean arterial pressure (MAP) during vehicle, phenylephrine (Phenyl) and angiotensin II (A-II) infusions from the first study. Changes (A) in pressure were comparable during phenylephrine and angiotensin II infusions and significantly greater than during vehicle infusion. Bottom, Response of forearm vascular resistance (FVR) during vehicle, phenylephrine (Phenyl) and angiotensin II (A-II) infusions from the first study. Only the increase during angiotensin II infusion was significant. The response during vehicle and phenylephrine infusions was not different. A = change.
Discussion Present studies. The results of these studies are consistent with the hypothesis that modestly pressor (first study) and nonpressor (second study) infusions of angiotensin II inhibit peripheral vascular responses to activation of baroreflexes in normal humans. In the first study, despite comparable elevation of mean arterial pressure by angiotensin II and phenylephrine, forearm vascular resistance did not decrease during angiotensin II hut actually increased. During systemic infusion of any pressor, forearm vascular responses would be the result of the balance between direct vasoconstriction and withdrawal of endogenous vascular tone as arterial pressure increases and baroreflex activation occurs. Because there was no change in forearm vascular resistance during phenylephrine, these effects presumably balanced. With angiotensin II, forearm vascular resistance increased markedly during the same change in mean
JACC Vol. 25, No. l January 1995:246-50
G O L D S M I T H AND HASKING ANGIOTENSIN It AND V A S C U L A R PRESSURE RESPONSE
249
Table 2. Responses to Vehicle and Angiotensin II Alone and During Head-Down Tilt and Head-Down Tilt With Phenylcphrine Infusion and Comparison With Vehicle in Eight Normal Subjects
Heart rate (beats/min) Vehicle Angiotensin II Mean arterial pressure (mm Hg) Vehicle Angiotensin II Forearm blood flow (ml/100 ml per min) Vehicle Angiotensin II Forearm vascular resistance (U) Vehicle Angiotensin I1 Norepinephrine spillover (ng/min) Vehicle Angiotensin II
Control
Infusion
HeadDown Tilt
Phenylephrineand Head-DownTilt
61 +_ 11 60 -+ 6.9
60 _+11 58 _+9.0
58 _+9.5 59 _+15
54 +_11" 51 _+7.6*
85 _+7.8 88 -+ 6.7
87 -+ 8.1 93 _+8.2
87 -+ 7.6 93 _+9.3
91 +_6.9* 102 ÷ 11'
1.8 _+0.27 2.4 _+0.78
2.1 _+0.62 2.4 _+0.89
2.1 _+0.58 3.0 _+1.3'
2.4 - 0.67¢ 2.7 -+ 0.99
48 _+9.3 41 -- 17
44 - 15 44 _+22
44 _ 12 37 _+ 18
40 _+l l t 43 _+21
612 ÷ 367 395 -- 105
637 +_387 451 _+283
548 +_265 518 _+445
418 + 196t 476 _+316
*p < 0.001, ?p < 0.05 versus control. Data presented are mean value _+SD.
arterial pressure. This result would be consistent with inhibition of baroreflex-mediated suppression of endogenous vasoconstrictor tone. In the second study, head-down tilt failed to produce a decrease in forearm vascular resistance or norepinephrine spillover on the vehicle day. This stimulus is predominantly to the cardiac reflexes because it increases central venous volume and pressure but not arterial pressure (8). When head-down tilt was combined with phenylephrine, arterial pressure increased, and heart rate, forearm vascular resistance and norepinephrine spillover all declined significantly, as expected during vehicle infusions. Heart rate also declined during angiotensin II, but the change in mean arterial pressure was greater when phenylephrine was added to the subpressor
Figure 2. Ratio of changes (A) in heart rate (HR) and mean arterial pressure (MAP) during head-down flit combined with phenylephrine infusion in the second study. As a function of change in mean arterial pressure, the decrease in heart rate was greater when the loading stimulus was applied during vehicle infusion than during angiotensin II (A-II) infusion.
angiotensin II than it was with vehicle. The change in heart rate relative to arterial pressure, although of borderline statistical significance between angiotensin II and vehicle days (Fig. 2), was less during angiotensin II, suggesting that as in previous studies (2), angiotensin II attenuated baroreflex-mediated responses to the increase in arterial pressure. Forearm vascular resistance declined appropriately when phenylephrine was given in the presence of the vehicle for angiotensin II but failed to decrease during phenylephrine plus angiotensin II despite the higher mean arterial pressure. In this case, the change in resistance relative to pressure was markedly less with angiotensin II than with vehicle (Fig. 3). The response of norepinephrine spillover (Table 2) paralleled that of forearm vascular
Figure 3. Ratio of the changes (A) in forearm vascular resistance (FVR) and mean arterial pressure (MAP) during head-down tilt combined with phenylephrine infusion in the second study. As a function of change in mean arterial pressure, the decrease in forearm vascular resistance was greater when the loading stimulus was applied during vehicle infusion than during angiotensin II (A-II) infusion.
-1 -1
-2 AFVRIAMAP
AHR/AMAP -2
-3 ,.J
-3 *p<.07 -4
-5
*p<.OS
-6 Vehicle
A-II
Vehicle
A-II
250
GOLDSMITH AND HASKING ANGIOTENSIN II AND VASCULAR PRESSURE RESPONSE
resistance, with a significant decline during head-down tilt and phenylephrine in the presence of vehicle and no change in the presence of angiotensin II despite a higher pressure. It is therefore likely that in these studies, angiotensin II attenuated the heart rate and the peripheral vascular and systemic sympathetic aspects of the efferent limb of the baroreflex. We should note that we have not performed a complete test of baroreflex gain and set point because we have only one interventional time point, but because pressure began at similar levels before phenylephrine on each study day, it is unlikely that we began at markedly different positions on the operating curve for the baroreceptors, making our comparison of the ratios of heart rate and forearm resistance to pressure a reasonable one. A more complete study over a wide range of pressure would be required to determine whether the effect of angiotensin II was on the gain or threshold of baroreflex function. Previous studies. These results are consistent with an observation by Matsukawa et al. (4), who reported attenuation of the muscle sympathetic nerve response to increased arterial pressure induced by angiotensin II relative to phenylephrine. Those investigators found a profound difference in muscle sympathetic nerve activity suppression after equipressor infusions of angiotensin II and phenylephrine yet nearly no effect of angiotensin II on muscle sympathetic nerve activity when blood pressure was held constant with nitroprusside. The latter result is consistent with our previous and current observations regarding the absence of effects of subpressor angiotensin II infusions on systemic norepinephrine spillover. There is, therefore, little chance that a direct stimulation of sympathetic activity by angiotensin II at these infusion rates confounded our interpretation of the changes during baroreflex activation. Our data combined with those of Matsukawa et al. suggest an inhibitory effect of angiotensin II on heart rate and systemic sympathetic and peripheral vascular responses to baroreceptor loading in normal humans. The effect on heart rate and sympathetic responses seems clear, but it remains possible that at least part of the attenuation of the forearm vascular responses could have been a result of amplification by angiotensin II of the effects of the phenylephrine in the forearm. However, two arguments may be made against this possibility. First, if such an effect occurred and contributed to the higher arterial pressure with an intact baroreflex, systemic norepinephrine spillover should have declined. Yet norepinephrine spillover declined when pressure increased with phenylephrine plus vehicle but did not decrease when angiotensin II was added to phenylephrine despite the higher pressure (Table 2). Second, in a previous study using 2 ng/kg per min infusions of angiotensin II (3), forearm vascular resistance responses were not potentiated during upright tilt in the presence of angiotensin II compared with vehicle. In that study, during upright tilt with vehicle infusions, forearm vascular resistance increased from 39 _+ 10 to 52 _+ 17 U (p < 0.05), whereas in the presence of 2 ng/kg per min of angiotensin II, it increased from 58 _+ 16 to 68 _+ 23 U (p < 0.05). The increase in forearm vascular resistance during upright tilt was no different between days.
JACC gol. 25, No. 1 January 1995:246-50
There is no obvious reason why angiotensin II would amplify an exogenous alpha-adrenergic stimulus (phenylephrine) in the forearm circulation during head-down tilt but not amplify an endogenous alpha-adrenergic stimulus in the forearm during the relatively potent sympathetic stimulation offered by upright tilt. The most likely explanation for the difference in forearm vascular responses is, therefore, baroreflex inhibition if neither direct sympathostimulation nor interaction with phenylephrine is probable. The results of these studies are in complete accord with several studies suggesting that angiotensin II inhibits baroreflex responses to baroreceptor loading in experimental animals (6). The most relevant single study is probably by Guo and Abboud (9), who documented that angiotensin II inhibited cardiac and peripheral sympathetic responses to an increase in pressure but did not restrain these responses during hypotension. Because, as noted, we also found no effect of either subpressor or pressor angiotensin II infusions on heart rate or systemic sympathetic or forearm vascular responses to baroreflex unloading with head-up tilt, it seems that angiotensin II interacts with the baroreflex only when pressure increases. Not resolved in any species is whether angiotensin II can inhibit the response to normotensive loading of the cardiopulmonary baroreceptors. Summary. Previous and present data do not support a direct effect of systemic infusions of angiotensin II on sympathetic activity in normal humans. Systemic infusion of angiotensin II is also without apparent direct effect on systemic norepinephrine spillover in congestive heart failure (10). However, baroreflex inhibitory effects of angiotensin II could contribute significantly to sympathetic dysregulation in hypertension and heart failure, representing an indirect mechanism by which angiotensin II contributes to volume expansion and excess vasoconstriction in these diseases.
References 1. Mace PJE, Watson RDS, Skan W, Littler WA. Inhibition of the baroreceptor heart rate reflex by angiotensin II in normal man. Cardiovasc Res 1985;19: 525-7. 2. Goldsmith SR, Hasking GJ. Effect of a pressor infusion of angiotensin II on sympathetic activity and heart rate in normal humans. Circ Res 1991;68: 263-8. 3. Goldsmith SR, Hasking GJ. Subpressor infusions of angiotensin II do not stimulate sympathetic activity in humans. Am J Physiol 1990;258:H179-82. 4. Matsukawa T, Gotoh E, Minimijawa K, et aL Effects of intravenous infusions of angiotensin II on muscle sympathetic nerve activity in humans. Am J Physiol 1991;261:690-6. 5. James TN. Absence of direct chronotropic action of angiotensin infused into the sinus node artery. Am J Physiol 1965;209:571-6. 6. Reid 1. Interactions between angiotensin II, sympathetic nervous system, and baroreceptor reflexes in regulation of blood pressure. Am J Physiol 1992; 262:E763 78. 7. Esler M, Jackman G, Bobik A, et al. Determination of norepinephrine apparent spillover rate and clearance in humans. Life Sci 1979;25:1461-70. 8. Goldsmith SR, Cowley AW, Francis GS, Cohn JN. Effects of increased central venous and aortic pressure on plasma vasopressin in man. Am J Physiol 1984;246:H647-51. 9. Guo GB, Abboud FM. Angiotensin II attenuates baroreflex control of heart rate and sympathetic activity. Am J Physiol 1984;246:H80-9. 10. Goldsmith SR, Hasking GJ. Angiotensin-II and sympathetic nervous activity in congestive heart failure. J Am Coil Cardiol 1993;21:1107-13.