Regional Hemodynamic and Endocrine Effects of Aldosterone and Cortisol in Conscious Sheep
Comparison With the Effects of Corticotropin
Abstract We studied the cardiovascular responses to 5 days’ infusion of aldosterone (10 μg/h) and cortisol (5 mg/h) to determine the possible contribution of mineralocorticoid and glucocorticoid actions to the regional hemodynamic changes caused by corticotropin. These infusion rates produce plasma levels similar to those seen during corticotropin stimulation. In five conscious sheep aldosterone progressively increased mean arterial pressure (P<.001) to a maximum of 11 mm Hg on day 5, whereas cortisol increased pressure by 5 mm Hg (P<.01) within 24 hours. Cardiac outputs on the control day and on day 5 of infusion were 4.4±0.3 and 4.9±0.3 L/min, respectively, for aldosterone and 4.3±0.4 and 5.0±0.4 L/min for cortisol. Neither steroid significantly altered total peripheral conductance, but they had different, nonuniform regional hemodynamic effects. Mesenteric conductance fell progressively with aldosterone from 7.14±0.35 (mL/min)/mm Hg to a minimum of 6.17±0.38 (P<.01) on day 5 of infusion. Mesenteric conductance was transiently reduced with cortisol, but this was not significant over the 5 days. Renal conductance was unchanged with aldosterone, but cortisol caused a rapid, sustained increase in renal conductance from 2.9±0.3 to 4.0±0.4 (mL/min)/mm Hg (P<.001) within 24 hours, similar to the increase caused by corticotropin. As with corticotropin there were only minor changes in the coronary and iliac vascular beds. In summary, these two endogenous steroids had contrasting, nonuniform regional hemodynamic effects, aldosterone causing mesenteric vasoconstriction, and cortisol causing renal vasodilatation. These findings suggest that corticotropin-induced renal vasodilatation is caused by the cortisol released by corticotropin and the mesenteric vasoconstriction partly depends on the mineralocorticoid actions of the adrenocortical steroids released by corticotropin.
- cardiac output
- hypertension, experimental
- splanchnic circulation
- renal circulation
Infusion of corticotropin causes a sustained, adrenally dependent hypertension. We have recently demonstrated that in conscious sheep corticotropin infusion initially increased CO, but arterial pressure was not elevated because of a rapid increase in renal conductance over the first 6 hours.1 After 14 hours of corticotropin infusion mesenteric conductance increased, and from this time arterial pressure began to increase. The increases in arterial pressure, CO, and renal conductance and the fall in mesenteric conductance were maintained over 5 days of corticotropin infusion.
The pressor effect of corticotropin in conscious sheep can be mimicked by infusion of a combination of seven adrenocortical steroids.2 3 The aim of the present study was to examine the regional hemodynamic effects of aldosterone and cortisol to determine the extent to which mineralocorticoid and glucocorticoid actions could account for the hemodynamic changes that occur with infusion of the seven combined steroids and by analogy corticotropin. The infusion rates of cortisol (5 mg/h) and aldosterone (10 μg/h) that were used have glucocorticoid and mineralocorticoid potencies, respectively, similar to those of the combined steroid infusion.3 The cortisol infusion rate was the same as that in the combination of steroids, which reproduces the pressor action of corticotropin.2 This infusion rate gives plasma cortisol levels similar to those achieved by maximal stimulation of the adrenal cortex with corticotropin.4 The aldosterone infusion rate produces plasma levels similar to those present on the first day of corticotropin treatment, although after 3 days of corticotropin, plasma aldosterone has returned to control.5
Previous studies have demonstrated that these two classes of steroids raise blood pressure by different mechanisms. The onset of mineralocorticoid hypertension is slow, developing over several days or weeks, and is associated with volume expansion caused by salt and water retention. The hypertension is potentiated by a reduction in renal mass and is related to sodium intake.6 7 8 Initially, CO is increased, but over several weeks CO returns to control levels and the hypertension is maintained by a secondary vasoconstriction.9 10 11 In contrast, glucocorticoid hypertension develops relatively rapidly; it is characterized by an independence from dietary Na+ intake and a shift of fluid from the interstitial to plasma volume compartment.7 12 In humans cortisol-induced hypertension is associated with an increase in CO with little change in total peripheral resistance.13
Although the effects of mineralocorticoids and glucocorticoids on blood pressure and CO have been documented, this study gives new information on the hemodynamic effects of these steroids in the major regional vascular beds, as well as indicating their possible involvement in the regional hemodynamic actions of corticotropin. We also studied the influence of these steroids on sodium and water balance and on plasma levels of ANP, ET-1, and PRC.
Merino cross ewes (35 to 45 kg body weight), oophorectomized and with carotid artery loops, were housed in individual metabolic cages in association with other sheep in an open laboratory. They were not used for at least 4 weeks after surgery until they were accustomed to laboratory conditions and human contact. Sheep were fed a diet of oaten chaff (800 g/d containing 90 to 120 mmol/kg Na+ and 270 to 380 mmol/kg K+), and water was offered ad libitum. All experiments were approved by the Animal Experimentation Ethics Committee of the Howard Florey Institute.
Animal Instrumentation and Data Collection
Flow probes were implanted in sheep for measurement of CO and regional flows as previously described.14 Briefly, anesthesia was induced with intravenous sodium thiopental (15 mg/kg). After intubation, sheep were placed on a ventilator and maintained on 1.5% halothane in air and oxygen. An electromagnetic flow probe (EMF; 20 mm diameter, In Vivo Metrics) was implanted on the ascending aorta, and a transit-time flow probe (3 mm RS with coronary flange, Transonic Systems Inc) was implanted on the left circumflex coronary artery. Two weeks later, transit-time flow probes were implanted on the cranial mesenteric artery (6 mm RS), the left renal artery (4 mm RS), and the external iliac artery (6 mm RS).
The transit-time flow probes were connected to a Transonic T201CDS flowmeter via a four-channel sequential scanner (TM04, Transonic), and the EMF probes were activated by a Biotronex flowmeter. The output voltage of the EMF meter was reset to zero with the use of an autozero circuit during a portion of each diastole when blood flow in the ascending aorta is assumed to be zero.14 The autozero circuit also incorporated a separate circuit to measure the first differential of the upstroke of systole (dF/dt) at each beat. A month after implantation, the EMF probes were calibrated in vivo against thermodilution over a range of CO values.14 Dobutamine (Dobutrex, Eli Lilly & Co) was used to increase CO from approximately 4 to 9 L/min.
Blood pressure was measured via an indwelling Tygon cannula (0.1 mm ID, 1.5 mm OD) inserted 15 cm into a carotid artery. The cannula was connected to a pressure transducer (TDXIII, Cobe) tied to the wool on the sheep’s back. The signal from the transducer was amplified and the output voltage corrected to compensate for the height of the transducer above heart level.14 The transducer was calibrated daily against a mercury manometer.
CVP was measured via a polyethylene cannula (1.18 mm ID, 1.7 mm OD) inserted 25 cm into a jugular vein connected to a pressure transducer (TDXIII, Cobe) and was recorded on a chart recorder (RS 3400, Gould Instruments). CVP was measured at about 10 am every day for 15 minutes with the sheep standing quietly with their heads in a neutral position. The transducer was positioned at heart level and was calibrated against a water manometer.
Analog signals (blood pressure, CO, dF/dt, and regional flows) were collected with a PC 486 data-acquisition system with custom software. After analog-to-digital conversion, data were collected at 100 Hz for 10 seconds at 10-minute intervals. The following cardiovascular variables were recorded: MAP, HR (calculated from the CO signal), CO, dF/dt, peak aortic flow (Fmax), SV (CO/HR), TPC (CO/MAP), mean regional blood flows, and mean regional conductances (mean flow/MAP). Cardiovascular variables were monitored from 10 am on the first day of the experiment to 9:50 am on the last day of the experiment in five sheep. Individual data points were pooled into 24-hour means for day-to-day comparisons.
After a control day, aldosterone (10 μg/h), cortisol (5 mg/h), or 0.15 mol/L NaCl (1.2 mL/h) was infused from 10 am on the first day of infusion for 5 days, followed by 3 postinfusion days. For all sheep, daily measurements of water intake, urine volume, urinary Na+ excretion, and urinary K+ excretion were performed at about 10 am. Carotid arterial blood samples (16 mL) for the determination of hematocrit, plasma Na+, K+, osmolality, total protein, glucose, ANP, ET-1, and PRC were taken at 10 am immediately before the onset of infusion after 1, 3, and 5 days of infusion as well as 3 days after infusion. Infusions were given in a random order with a minimum of 2 weeks between experiments.
Steroids (obtained from Steraloids) were kept in stock solutions of 100% ethanol, which were diluted daily with 5% dextrose and infused at a rate of 10 mL/h. The final ethanol concentration was 2% for aldosterone and 6% for cortisol.
With the use of an identical protocol, the effect of 5 days of intravenous infusion of vehicle (6% ethanol in 5% glucose infused at 10 mL/h) was examined in five noninstrumented conscious sheep. Vehicle infusion did not change MAP and HR; on the control and fifth infusion days MAP was 85±3 and 84±2 mm Hg and HR was 66±5 and 65±4 beats per minute, respectively. Vehicle infusion had no effect on plasma levels of Na+, K+, total protein, and osmolality. Hematocrit, water intake, urine volume, and 24-hour urinary Na+ and K+ excretions were also unchanged.
Plasma and Urine Analyses
Hematocrit was measured with a microhematocrit centrifuge (Biofuge A, Heraeus Sepatech). Plasma and urinary Na+ and K+ and plasma glucose were measured with a Synchron CX-5 Clinical System (Beckman Instruments). Plasma osmolality was measured with an osmometer (model 3CII, Advanced Instruments Inc).
Plasma ANP and ET-1 were assayed by radioimmunoassay after extraction from plasma (2 to 3 mL) with the use of Sep-Pak columns. The detection limit of the ANP assay was 0.8 fmol/mL of plasma, and the intra-assay and interassay coefficients of variation were 9% and 14%, respectively. For the plasma ET-1 assay the detection limit was 0.8 fmol/mL of plasma, and the intra-assay and interassay coefficients of variation were 9% and 12%, respectively. PRC was measured with an antibody capture technique.
Cardiovascular parameters (pooled into 24-hour means), plasma and urinary parameters, and water intake were compared by repeated-measures ANOVA with the Greenhouse-Geiser correction (comparing the pretreatment day and experimental days 1, 3, and 5 with saline control). Changes in plasma hormone levels and CVP were assessed by comparing the preinfusion value with values on experimental days 1, 3, and 5 by two-way ANOVA. Significant changes were accepted at a value of P<.05.
Aldosterone infusion (10 μg/h) caused a progressive increase in MAP that reached a maximum of 11 mm Hg above control on treatment day 5 (P<.001, Fig 1A⇓). CVP, measured in four of the sheep, was elevated by 2.1±0.5, 2.3±0.5, and 2.1±0.5 mm Hg (P<.01) on days 1, 3, and 5 of aldosterone, respectively. On the first postinfusion day CVP had returned to control levels. HR, Fmax, and dF/dt did not change, but SV was significantly increased (P<.05). CO was 4.40±0.34 L/min on the control day, 4.97±0.45 on the third infusion day, and 4.91±0.34 on the fifth infusion day, but these changes were not significantly different from control. TPC was unchanged, but aldosterone caused a progressive reduction in mesenteric conductance over the infusion period (P<.01, Fig 2A⇓). Coronary, renal, and iliac conductances remained unchanged. Renal blood flow increased from a control value of 258±22 mL/min to a maximum of 318±24 on treatment day 5 (P<.001, Fig 3A⇓). Coronary, mesenteric, and iliac flows were not altered by aldosterone infusion.
In contrast to aldosterone, the hemodynamic effects of cortisol (5 mg/h) were rapid in onset, reaching a plateau within 24 hours of the start of infusion. MAP was elevated by 5 mm Hg within 24 hours of cortisol infusion and remained at this level for the 5 days of treatment (P<.01, Fig 1B⇑). During cortisol infusion CVP, measured in three of the sheep, was unchanged (1.0±0.6, 0.5±0.5, and 0.5±0.5 mm Hg above control on days 1, 3, and 5 of infusion, respectively). On the control day CO was 4.28±0.35 L/min, and on the first and fifth infusion days CO values were 4.84±0.36 and 4.99±0.42 L/min, respectively, but these increases were not significantly different from control. HR and SV did not change significantly, whereas both Fmax and dF/dt were increased (both P<.05). TPC was unchanged, but there were differential changes in individual vascular beds. Cortisol caused a large, rapid increase in renal conductance from 2.85±0.32 (mL/min)/mm Hg at 10 am, when cortisol infusion started, to 3.88±0.39 after 5 hours and 4.45±0.48 after 10 hours of infusion. Renal conductance remained elevated at this level for the 5 days of infusion (P<.001, Fig 2B⇑). Mesenteric conductance initially fell and then returned to preinfusion levels by day 4 of cortisol, but these changes were not significant over the 5 infusion days. Coronary and iliac conductances remained unchanged during cortisol infusion. Renal flow was increased throughout the cortisol infusion period by a maximum of 53% on the second infusion day (P<.001), but coronary, mesenteric, and iliac flows were unaltered (Fig 3B⇑). Saline infusion had no effect on any of the variables measured.
Throughout the aldosterone infusion, plasma [Na+] was increased (P<.05) and plasma [K+] reduced (P<.001, Table 1⇓). Hematocrit, plasma glucose, osmolality, and water intake did not change. Urinary Na+ excretion was initially reduced (P<.01) but returned to preinfusion levels by the end of the infusion. Urinary K+ excretion remained unchanged. Diuresis was associated with natriuresis on the first postinfusion day, but urine volume did not change significantly over the 5 days of aldosterone infusion.
Cortisol infusion increased plasma glucose (P<.001) but had no effect on osmolality or hematocrit (Table 2⇓). There was a small decrease in plasma [Na+] (P<.05) but no effect on plasma [K+]. Water intake did not change, although urine volume increased throughout the cortisol infusion period (P<.01) and on the first postinfusion day. Urinary Na+ excretion increased during the latter days of cortisol infusion (P<.05), and urinary K+ excretion remained unchanged.
PRC was suppressed to undetectable levels by aldosterone infusion (P<.05) but was not changed by cortisol infusion (Table 3⇓). Plasma ANP was 3.49±0.58 pmol/L on the control day and 11.85±3.91 pmol/L on day 5 of aldosterone infusion, but this difference was not significant because of the large variability caused by a lack of a rise in ANP in one sheep. Cortisol infusion did not alter ANP levels. Neither steroid altered plasma ET-1 concentration.
In this study we examined the cardiovascular effects of infusion of the major ovine glucocorticoid cortisol and the major ovine mineralocorticoid aldosterone in conscious sheep. The infusion rates used produce plasma levels similar to those reached during corticotropin infusion.2 4 5 Aldosterone caused hypernatremia, hypokalemia, and initially, urinary Na+ retention, indicative of mineralocorticoid activity. Cortisol infusion was associated with the glucocorticoid effects of hyperglycemia and increased urine volume.
Aldosterone caused a progressive increase in MAP, which reached 11 mm Hg above control on the fifth treatment day, an increase similar to that previously reported by this laboratory.3 The rise in MAP resulted mainly from an increase in CO caused by a significant increase in SV. Both indexes of left ventricular contractility were unchanged, suggesting that the rise in SV resulted from the increased preload, as indicated by the sustained elevation in CVP. The rise in CVP was probably secondary to the volume expansion caused by mineralocorticoid-induced urinary Na+ retention.
Similar hemodynamic changes occurred during the initial phase of hypertension in patients with primary aldosteronism, in whom hypertension developed after withdrawal of spironolactone treatment,9 10 and in dogs infused with aldosterone.11 In these cases the elevated CO was also mediated by an increase in SV caused by an expanded plasma volume. A switch to a resistance driven hypertension, which has been reported by other authors,9 10 11 was not seen in this relatively short-term study.
At the infusion rate used, cortisol caused approximately half the increase in MAP seen with aldosterone, although the increases in CO were similar. The lower increase in MAP with cortisol resulted from the tendency for TPC to increase during cortisol infusion (+13±5% by day 5), whereas at the end of the aldosterone infusion TPC was at control levels. The rise in TPC during cortisol indicated that, as in humans,13 the increase in MAP depended on the increase in CO. As with aldosterone, the increase in CO with cortisol depended on a rise in SV, which with both steroids was about 8 mL per beat. This increase in SV with cortisol appeared to be accounted for mainly by an increase in left ventricular contractility, because CVP did not increase. These changes with aldosterone and cortisol are qualitatively similar to those that occurred with corticotropin infusion.1 Corticotropin infusion caused a greater increase in MAP, accompanied by an increase in CO that resulted from a rise in SV caused by increases in CVP and ventricular contractility.
The regional hemodynamic responses to these two endogenous steroids were clearly different, and the effects were nonuniform across the vascular beds studied. Mesenteric conductance fell progressively during aldosterone infusion; by the end of the infusion the decrease (−13±2%) was similar to that seen with corticotropin (−11±6%).1 The aldosterone-induced mesenteric vasoconstriction, which may be accompanied by vasoconstriction in the celiac bed as was found with corticotropin, probably accounted for the fall in TPC and increase in blood pressure over the last 3 days of infusion when CO was unchanged. In contrast, cortisol caused a nonsignificant, transient fall in mesenteric conductance. These findings suggest that the mineralocorticoid action of the adrenocortical steroids released by corticotropin could progressively account for the fall in mesenteric conductance during corticotropin infusion, whereas in the initial phase other mechanisms are involved.
The mechanisms by which aldosterone causes mesenteric vasoconstriction are unknown. There is evidence that increased ANP levels may selectively cause mesenteric vasoconstriction15 and that pressor sensitivity to norepinephrine and angiotensin is increased during infusion of fludrocortisone16 and deoxycorticosterone acetate17 in humans, but whether these effects occur in sheep and whether they are selective to the mesenteric vascular bed are unknown.
The increases in renal blood flow and renal conductance after intravenous cortisol infusion are in agreement with previous findings in sheep.18 Numerous studies have described increased renal blood flow and increased glomerular filtration rate as a consequence of glucocorticoid administration, but the mechanisms involved remain unknown.19 20 The time course and degree of renal vasodilatation with cortisol (+32±7% on day 5) were similar to that caused by corticotropin (27±7%),1 indicating that this effect was mediated by the cortisol secreted in response to corticotropin. Aldosterone had no effect on renal conductance, but during the infusion there was a progressive increase in renal blood flow that was presumably secondary to the increase in blood pressure.
The combined metabolic changes caused by aldosterone and cortisol can account for many of the changes reported with corticotropin.1 The cortisol-induced increases in plasma glucose concentration and urine volume were similar to the increases seen with corticotropin. Aldosterone and corticotropin caused a similar degree of hypokalemia and urinary Na+ retention, and with both treatments there was a similar postinfusion natriuresis.
The endocrine changes caused by these two steroids reflected their cardiovascular and metabolic actions. The aldosterone-induced sodium retention and volume expansion caused a large fall in PRC to undetectable levels, as is found in patients with primary aldosteronism and in dogs infused with aldosterone.10 21 In contrast, cortisol caused a much smaller fall in PRC, analogous to the findings in dogs.22 The changes in ANP levels correlated with CVP and therefore atrial stretch, which is known to be a stimulus for ANP release. Corticotropin, which caused a larger increase in CVP, caused a greater rise in ANP.1 Aldosterone did not change ET-1 levels, in agreement with a similar finding in patients with primary aldosteronism23 and during corticotropin infusion in sheep.1
In summary, aldosterone infusion (10 μg/h) into conscious sheep caused approximately double the increase in MAP compared with cortisol infusion (5 mg/h), although the increases in CO were similar with both steroids. Each steroid caused characteristic regional hemodynamic changes: cortisol caused renal vasodilatation, and aldosterone caused mesenteric vasoconstriction. These findings suggest that the renal vasodilatation that occurs during corticotropin infusion is a glucocorticoid effect, whereas the mesenteric vasoconstriction depends partly on the mineralocorticoid actions of the adrenocortical steroids released by corticotropin.
Selected Abbreviations and Acronyms
|ANP||=||atrial natriuretic peptide|
|CVP||=||central venous pressure|
|MAP||=||mean arterial pressure|
|PRC||=||plasma renin concentration|
|TPC||=||total peripheral conductance|
This work was supported by an institute grant to the Howard Florey Institute from the National Health and Medical Research Council. The authors are grateful to Tony Dornam for technical assistance, Rod Patterson for surgical assistance, Elizabeth Cooper for the assay of plasma ANP and ET-1, and Lisa Robinson for the measurement of PRC.
Reprint requests to Dr C.N. May, Howard Florey Institute, University of Melbourne, Parkville 3052, Australia. E-mail Clive_May.firstname.lastname@example.org.
- Received November 18, 1994.
- Revision received January 3, 1995.
- Accepted April 28, 1995.
Bednarik JA, May CN. Differential regional hemodynamic effects of corticotrophin in conscious sheep. Hypertension. 1994;24:49-55.
Fan JSK, Coghlan JP, Denton DA, Oddie CJ, Scoggins BA, Schulkes AA. Effect of intravenous infusion of corticosteroids on blood pressure, electrolytes and water metabolism in sheep. Am J Physiol. 1975;228:1695-1701.
Scoggins BA, Coghlan JP, Denton DA. ACTH-induced hypertension in sheep. In: de Jong W, ed. Handbook of Hypertension, Volume 4: Experimental and Genetic Models of Hypertension. Amsterdam, Netherlands: Elsevier Science Publishers; 1984:107-134.
Friedman SM, Friedman CL, Polley JR. Potentiation of hypertensive effects of DCA by various sodium salts. Am J Physiol. 1948;153:226-234.
Knowlton AI, Loeb EN, Stoerk HC, White JP, Heffernan JF. Induction of arterial hypertension in normal and adrenalectomized rats given cortisone acetate. J Exp Med. 1952;96:187-205.
Montani J-P, Mizelle L, Adair TH, Guyton AC. Regulation of cardiac output during aldosterone-induced hypertension. J Hypertens. 1989;7:S206-S207.
Haag D, Mohring B, Petri M, Hackenthal E. Comparative study on development of corticosterone and DOCA hypertension in rats. Am J Physiol. 1977;233:F403-F411.
Sudhir K, Jennings GJ, Esler MD, Korner PI, Blombery PA, Lambert GW, Scoggins BA, Whitworth JA. Hydrocortisone-induced hypertension in humans: pressor responsiveness and sympathetic function. Hypertension. 1989;13:416-421.
Bednarik JA, May CN. Evaluation of a transit-time system for the chronic measurement of regional blood flow in conscious sheep. J Appl Physiol. 1995;78:524-530.
Woods RL, Smolich JJ. Regional blood flow effects of ANP in conscious dogs: preferential gastrointestinal vasoconstriction. Am J Physiol. 1991;261:H1961-H1969.
Raab W, Humphreys RJ, Lepeschkin E. Potentiation of pressor effects of norepinephrine and epinephrine in man by deoxycorticosterone acetate. J Clin Invest. 1950;29:1397-1404.
McDougall JG, Butkus A, Coghlan JP, Denton DA, Meehan PJ, Potocnik SJ, Scoggins BA, Wright RD. Dissociation of biological effects and classical receptor binding of C-21 steroids in the kidney. In: Mantero F, Biglieri EG, Funder JW, Scoggins BA, eds. The Adrenal Gland and Hypertension. New York, NY: Raven Press Publishers; 1985:303-306.
Hall JE, Morse CL, Smith MJ, Young DB, Guyton AC. Control of arterial pressure and renal function during chronic glucocorticoid excess in dogs. Hypertension. 1980;2:139-148.
Pan Y-J, Young DB. Experimental aldosterone hypertension in the dog. Hypertension. 1982;4:279-287.
Lohmeier TE, Kastner PR. Chronic effects of ACTH and cortisol excess on arterial pressure in normotensive and hypertensive dogs. Hypertension. 1982;4:652-661.
Veglio F, Melchio R, Rabbin F, Chiandussi L. Plasma immunoreactive endothelin-1 in primary aldosteronism. Am J Hypertens. 1994;7:559-561.