Patients with primary aldosteronism (PA) were shown to have suppressed muscle sympathetic nerve activity (MSNA) in our previous study. Although baroreflex inhibition probably accounts in part for this reduced MSNA in PA, we hypothesized that the lowered activity of the renin-angiotensin system in PA may also contribute to the suppressed SNA. We recorded MSNA in 9 PA and 16 age-matched normotensive controls (NC). In PA, the resting mean blood pressure (MBP) and serum sodium concentrations were increased, and MSNA was reduced. We examined the effects of infusion of a high physiological dose of ANG II (5.0 ng·kg−1·min−1) on MSNA in 6 of 9 PA and 9 of 16 NC. Infusion of ANG II caused a greater pressor response in PA than NC, but, in spite of the greater increase in pressure, MSNA increased in PA, whereas it decreased in NC. Simultaneous infusion of nitroprusside and ANG II, to maintain central venous pressure at the baseline level and reduce the elevation in MBP induced by ANG II, caused significantly greater increases in MSNA in PA than in NC. Baroreflex sensitivity of heart rate, estimated during phenylephrine infusions, was reduced in PA, but baroreflex sensitivity of MSNA was unchanged in PA compared with NC. All the abnormalities in PA were eliminated following unilateral adrenalectomy. In conclusion, the suppressed SNA in PA depends in part on the low level of ANG II in these patients.
- angiotensin II
it is unclear whether or not a neurogenic mechanism is involved in the development and maintenance of elevated blood pressure levels in cases of hypertension induced by mineralocorticoid (2, 4, 5, 9, 17, 26, 29, 35). Previous studies in rats (2, 4, 5, 29) demonstrated the importance of the sympathetic nervous system or the central nervous system in the pathogenesis of hypertension caused by an excess of mineralocoticoid; central administration of aldosterone caused hypertension in rats (4, 5). However, recent studies in sheep did not support a role of the brain or the sympathetic nervous system in blood pressure elevation induced by the administration of aldosterone (17, 35). Some clinical studies have estimated sympathetic nerve activity during the administration of mineralocorticoid (9, 26) or the nerve activity in patients with primary aldosteronism (PA), a type of human mineralocoticoid hypertension (1, 8, 18, 20). Previous clinical studies, including a direct recording of sympathetic nerve activity in our previous study (18) and power spectral analysis of blood pressure (20), suggested a decrease in sympathetic nerve activity in PA patients. However, no studies on directly recorded sympathetic nerve activity in PA patients exist, or the mechanisms of the abnormality of nerve activity in PA were not clarified.
Circulating ANG II has been reported to increase sympathetic nerve activity in animals (24, 27) and humans (11, 14). Previous animal studies suggest that ANG II stimulates sympathetic nerve activity via sympathetic ganglia (3, 10, 11) and the central nervous system (24, 27). In our previous human studies with normal subjects, directly recorded sympathetic nerve activity increased while the pressor effects of ANG II were offset by simultaneous infusions of nitroprusside (13, 14). In addition, our previous studies suggested the occurrence of augmented sympathetic nerve activity in both patients with accelerated hypertension (15) and those with renovascular hypertension (18), which showed elevated renin-angiotensin system activity. In PA, renin-angiotensin system activity is known to be suppressed, but the effect of this suppressed activity on sympathetic neural outflow is unknown. We hypothesized that, in PA, lowered renin-angiotensin system activity contributed to the suppressed sympathetic nerve activity observed in patients with PA and carried out studies to address this hypothesis.
SUBJECTS AND METHODS
The subjects were 9 patients with PA (4 females and 5 males; average age 46.6 ± 4.1 yr) and 16 healthy volunteers as normal control subjects (7 females and 9 males; average age 47.9 ± 3.4 yr). The diagnosis of PA was based on the demonstration of increased concentrations of plasma aldosterone, suppressed plasma renin activity, and the existence of unilateral adrenal adenoma by computed tomography and adrenal scintillation scanning. The normal control subjects were within 10% of their ideal body weight and were shown to be free from endocrine, metabolic, and cardiovascular disorders by preliminary examinations. Serum concentrations of sodium were determined before the study. All subjects were hospitalized, received a diet containing 110 meq/day of sodium chloride, and took no medications for at least 2 wk before the study. All patients underwent surgery for unilateral adrenalectomy after monitoring muscle sympathetic nerve activity (MSNA) and received no medication after the surgery. After unilateral adrenalectomy, the PA patients were then subjected to the same protocol for recording MSNA between the 28th and 42nd day in eight patients and on the 88th day in one patient. Written informed consent was obtained from each subject following a detailed explanation of the purpose of the study and the procedures. The study protocols were approved by the Committee on Human Research of Japan Woman's University.
Procedures, Measurements of MSNA, Plasma Renin Activity, Plasma Aldosterone Concentrations, and Study Protocols
Subjects were examined in the supine position throughout each study session. Blood pressure was measured every minute using an automatic sphygmomanometer (Nihon Colin BP-203NP), and heart rate was measured from the electrocardiogram. Central venous pressure was measured continuously by means of a catheter introduced in the right antecubital vein and advanced to the superior vena cava. MSNA was recorded continuously from a muscle fascicle in the tibial nerve at the popliteal fossa by a microneurographic method (19–21). MSNA recordings were full-wave rectified and integrated with a time constant of 0.1 s. For quantitative analysis, sympathetic bursts were identified by inspecting the integrated neurogram and were expressed as bursts per minute (13–15). In addition, mean amplitudes of the MSNA bursts every minute were obtained, and total MSNA was determined as the product of mean amplitude and bursts per minute according to the method of previous studies (13, 14). The baseline control total MSNA was taken as 100% before ANG II or phenylephrine infusions.
After resting in the supine position for 30 min, resting values of blood pressure, heart rate, and MSNA were recorded during a 15-min interval before the start of the drug infusion. To determine the resting level of plasma renin activity and plasma concentrations of aldosterone, venous blood samples were collected via the venous cannula in chilled tubes containing EDTA-2Na and promptly centrifuged. Using commercially available kits (plasma renin activity “SRL” kit and SPAC-S Aldosterone kit; SRL), the plasma concentrations of ANG I and aldosterone were determined by RIA. The assay was used to measure plasma renin activity by incubating plasma at 37°C for 90 min and calculating the difference between the ANG I concentration during incubation of plasma at 4°C and that generated during incubation at 37°C.
Study protocols are shown in Fig. 1. Intravenous infusions of a high physiological dose of ANG II (5.0 ng·kg−1·min−1) were administered for 15 min to 6 of 9 patients with PA and 9 of 16 normal control subjects, that is, not all patients and normal subjects were used for the ANG II infusion study. The dose of ANG II (5.0 ng·kg−1·min−1 for 15 min) was chosen because the dose was thought to be enough to elevate blood pressure in normotensive control subjects and to be highly physiological in accordance with our previous studies (13, 14). To explore the effect of infusion of ANG II on sympathetic nerve activity when the baroreflex-mediated inhibition of sympathetic nerve activity induced by the elevations in blood pressure and central venous pressure is reduced, we used the following protocol: 30 min after the first infusion of ANG II, the same dose of ANG II was infused in the same subjects for 15 min with a simultaneous infusion of nitroprusside (0.05–0.50 μg·kg−1·min−1) to offset the elevations in blood pressure and central venous pressure induced by the ANG II, i.e., to maintain the central venous pressure at the baseline level and to keep the blood pressure constant at a lower level compared with during the infusion of ANG II alone. Mean blood pressure, heart rate, and MSNA were measured before and during the infusions of ANG II. The means of these values during the last 5-min interval were established as steady-state data during the infusion of ANG II.
To evaluate baroreflex sensitivities of heart rate and of MSNA, intravenous infusions of phenylephrine (0.5 and 1.0 μg·kg−1·min−1) were administered for 15 min separated by 30-min intervals in 5 of the 9 patients with PA and 8 of the 16 normal control subjects. Mean blood pressure, heart rate, and MSNA were measured before and during the infusions of phenylephrine. We obtained the mean values of mean blood pressure, heart rate, and MSNA during the 5-min period just before each infusion as baseline values and those during the last 5-min period of each infusion as the steady-state data for the infusion according to a previous study (13).
Synthetic ANG II (Deliver, Tohaeiyoh-Yamanouchi, Japan) was diluted in physiological saline (0.9%) and infused at a rate of 0.02 ml·kg−1·min−1 via a venous catheter with the use of a constant infusion pump (STC-52103; Terumo, Tokyo, Japan).
Data are expressed as means ± SE and were analyzed by analysis of variance. For F ratios significant at <0.05, either an unpaired Student's t-test or a paired Student's t-test was applied to determine the significance of the differences between given pairs of means, that is, the data obtained in the normal subjects and the PA patients were analyzed by an unpaired t-test, and the data obtained before and after surgery were analyzed by paired t-test. The relationship between mean arterial pressure and associated heart rate and that between mean arterial pressure and associated MSNA before and during the infusions of phenylephrine (see Table 3) was assessed by linear regression analysis for the data points located in the response curve including baseline data. Statistical significance was defined as a P value <0.05.
Patients with PA had significantly higher resting levels of blood pressure, serum sodium, and plasma aldosterone concentrations compared with normal control subjects (Table 1). There were no significant differences in the heart rate between the groups. The resting levels of plasma renin activity and MSNA and were significantly lower in patients with PA than in the normal subjects (Table 1 and Figs. 2 and 3).
Effect of intravenous infusions of ANG II on mean blood pressure, heart rate, and MSNA.
The infusion of a high physiological dose of ANG II produced an increase in mean blood pressure, and the pressor response was significantly enhanced in patients with PA compared with the normal subjects. Both absolute (burst/min) and relative (%) values of MSNA increased significantly in patients with PA but decreased in the normal subjects (Table 2 and Figs. 3 and 4). Simultaneous infusion of nitroprusside with ANG II induced an increase in both absolute and relative values of MSNA in both the normal control subjects and PA patients, although mean blood pressure still increased in both groups. The increase in MSNA was augmented in PA patients compared with the normal subjects in spite of greater increases in mean blood pressure in PA patients (Table 2 and Figs. 3 and 4).
After unilateral adrenalectomy resting values and effect of intravenous infusions of ANG II on mean blood pressure, heart rate, and MSNA.
After removing the adrenal adenomas, blood pressure, serum sodium concentrations, and plasma aldosterone concentrations dropped significantly, whereas plasma renin activity and resting MSNA increased (Table 1 and Figs. 2 and 3). The pressor response to the ANG II infusion dropped significantly after removal of the adrenal adenomas. MSNA increased significantly in the patients with PA during the infusion of ANG II before the removal of the adrenal adenomas but decreased after the treatment (Table 2 and Figs. 3 and 4). Simultaneous infusion of nitroprusside with ANG II elevated MSNA in PA patients both before and after adrenalectomy, and the increase in MSNA was augmented in PA patients before the treatment compared with those after the treatment.
Baroreflex sensitivities of heart rate and MSNA, estimated by phenylephrine infusions.
The baroreflex sensitivity of the heart rate was significantly smaller in PA patients than in normal subjects and increased significantly in PA patients after unilateral adrenalectomy. There were no significant differences in the baroreflex sensitivity of MSNA between normal subjects and PA patients before unilateral adrenalectomy or in PA patients before and after unilateral adrenalectomy (Table 3).
Baseline MSNA values in Tables 2 and 3 tended to be smaller in PA patients than in normal controls (P < 0.10) but not to a significant extent. This lack of significance may well be a type 2 error, since the comparisons are between small groups of different subjects. Indeed, the baseline MSNA values in PA patients (n = 9) were significantly smaller those in normal controls (n = 16), as shown in Fig. 2.
In this study, we made four main observations. First, MSNA was suppressed in patients with PA compared with normal control subjects, which was likely to be due to baroreflex inhibition in response to their high blood pressure. Second, we observed in PA patients that intravenous infusion of a high physiological dose of ANG II increased MSNA and blood pressure, whereas in the normal control subjects ANG II decreased MSNA. Moreover, the simultaneous infusion of nitroprusside with ANG II induced an increase in MSNA in both PA patients and normal controls, and the increase in MSNA induced by ANG II with nitroprusside was markedly greater in PA patients. Third, baroreflex sensitivity of the heart rate was reduced in PA patients, but the baroreflex sensitivity of MSNA was not different in PA patients. Fourth, after unilateral adrenalectomy, the resting MSNA and the MSNA and blood pressure responses to the infusion of ANG II were normalized.
Previous studies of mineralocoticoid hypertension in rats suggest that central mechanisms play a role in the pathogenesis of blood pressure elevation (2, 4, 5, 29), but this was shown not to be the case in sheep (17, 35) or humans (1, 8, 9, 18, 20, 26). In previous studies, we found that directly measured MSNA was lower in patients with PA compared with both normal subjects and patients with essential hypertension (18). Furthermore, sympathetic nerve activity evaluated indirectly with power spectral analysis of blood pressure was shown to be lowered in patients with PA before surgery compared with those after unilateral adrenalectomy (20). To date, there have been no systematic studies to directly evaluate sympathetic nerve activity in patients with PA before and after unilateral adrenalectomy. In the present study, we found that sympathetic neural outflow was lower in patients with PA compared with the values after unilateral adrenalectomy or compared with normal control subjects. These findings suggest that sympathetic regulatory systems play a role in the maintenance of blood pressure in patients with PA after unilateral adrenalectomy, but not in the maintenance of hypertension in patients with PA before unilateral adrenalectomy.
The reasons why the sympathetic neural outflow is suppressed in patients with PA are not known exactly, but some possible explanations are described below. The blood pressure elevation induced by mineralocorticoid excess will cause the inhibition of sympathetic nerve activity via baroreceptor reflex mechanisms in patients with PA. The baroreflex sensitivity of MSNA was intact in patients with PA. Therefore, if blood pressure is lowered with a vasodilator, such as nitroprusside, in PA patients, MSNA is thought to return to normal or increase (36). In deoxycorticosterone acetate-salt hypertensive rats, as a mineralocorticoid hypertensive animal model, infusion of nitroprusside induced an increase in renal sympathetic nerve activity (36). Conversely, in rats, aldosterone excess caused hypertension via central nervous system mechanisms (4, 5), which is in disagreement with our present findings. Recent studies in sheep did not support a role for the brain, the sympathetic nervous system, or its humoral mechanism in blood pressure elevation induced by the administration of aldosterone (17, 35). In our previous (18) and present studies, we found that resting MSNA was suppressed in PA. Another possible explanation for suppressed sympathetic neural outflow in patients with PA is lowered intrinsic renin-angiotensin system activity.
We also found that an intravenous infusion of ANG II produced an increase in MSNA in PA patients, whereas it caused a decrease, presumably via baroreceptor reflexes, in normal control subjects, that is, both absolute changes in MSNA (beats/min) as well as the difference in the relative changes induced by ANG II infusions in MSNA (%) were enhanced in PA patients compared with normal control subjects and PA patients after unilateral adrenalectomy. Furthermore, simultaneous ANG II and nitroprusside infusions induced an increase in MSNA in normal subjects, in agreement with our previous studies (13, 14), as well as in PA patients, and the increase in MSNA was markedly higher in PA patients in the present study. This suggests that ANG II may have stimulatory effects on the sympathetic nervous system in humans. The enhanced stimulatory effect of ANG II on sympathetic neural outflow observed in PA patients was thought to overcome the inhibitory effect on sympathetic neural outflow via baroreceptor reflexes during the pressor response to this peptide. The present study, showing that the infusion of ANG II produced increases in MSNA in patients with PA, supports the conclusion that the intravenous infusion of ANG II stimulates sympathetic nerve activity in humans.
According to data expressed on an absolute basis (beats/min) in Table 2, absolute MSNA was not statistically significantly greater during the infusion of ANG II with nitroprusside in PA patients (50.0 beats/min) than in normal control subjects (41.1 beats/min), but this lack of significance may well be a type 2 error, since the comparisons are between small groups of different subjects. However, there was a significantly greater increase in MSNA during ANG II and nitroprusside infusion in PA patients (20.0 bursts/min) vs. normal controls (4.1 bursts/min) even though the increase in mean blood pressure in PA patients (10.0 mmHg) tended to be greater than in the normal controls (4.9 mmHg). We think that these findings show that ANG II has a stimulatory action on sympathetic nerve activity in PA patients.
The exaggerated pressor response to an intravenous infusion of ANG II was observed in the PA patients compared with the same patients after unilateral adrenalectomy and with normal control subjects. A possible explanation for the enhanced pressor response to ANG II observed in PA patients is an upregulation of ANG II receptors in the vasculature (31, 32) and in the central and peripheral components of the sympathetic nervous system (5, 34, 37) during administration of mineralocorticoid. Mineralocorticoid was shown to increase ANG II binding in the vasculature (31, 32) and in the brain (5, 34, 37), and then it induces an upregulation of ANG II receptors. Another possible explanation for the exaggerated pressor response to ANG II in PA patients is in part due to an elevation of serum sodium concentrations. The pressor and sympathetic responses to ANG II are well known to be enhanced by a high-salt intake (22). In addition, aldosterone has been shown to increase cerebrospinal fluid sodium concentration, and reducing this leads to a fall in blood pressure (23), further demonstrating the important role of increased brain sodium in mineralocorticoid hypertension. Moreover, the enhanced pressor response to ANG II in PA patients may also be due to a greater increase in sympathetic neural outflow.
Baroreflex sensitivity of the heart rate was reduced in PA patients compared with normal control subjects and the same patients after unilateral adrenalectomy, in agreement with previous studies (16, 30). Previous studies showed that intravenous infusion of aldosterone attenuated the baroreflex sensitivity of the heart rate (19, 33) and MSNA (19). However, in the present study, the baroreflex sensitivity of MSNA was preserved in PA patients. This disagreement may be due to the difference in methods; we examined the effect of chronic high levels of aldosterone on baroreflex sensitivity of MSNA in PA patients, and the previous study examined the effect of acute infusions of aldosterone on that in normal subjects (19). Previous animal studies (6, 30) and human studies (12) showed that there were dissociations between baroreflex sensitivity of the heart rate and that of sympathetic nerve activity. This dissociation may reflect differences in the baroreflex control of parasympathetic and sympathetic nerve activity, that is, baroreflex control of heart rate has been shown to primarily involve a parasympathetic mechanism, since the heart rate response to changes in blood pressure was not inhibited by propranolol but was abolished by pretreatment with atropine (25).
In the present study, we evaluated only baroreflex sensitivity of the high-pressure limb and did not examine the full range of pressures for which the baroreflex may buffer changes. Our findings suggest that reduced baroreflex sensitivity of the high blood pressure limb of the reflex may contribute to a reduced baroreceptor-mediated inhibition of MSNA in PA patients, and thus the maintenance of hypertension in these patients.
The sites of action of ANG II that affect the sympathetic neural outflow could not be determined from the present study, but evidence from previous studies has suggested the involvement of the central nervous system (7, 21, 24, 26, 27, 34, 37) or sympathetic ganglia (3, 10, 11). The effect of circulating ANG II on the sympathetic nervous system has been postulated to be mediated via circumventricular organs, including the area postrema or the subfornical organs, which are regions that have no blood-brain barrier (7, 21, 24, 26, 27, 34, 37). In addition, ANG II has been shown to stimulate sympathetic ganglionic neurons (3, 10, 11).
In this study, we observed that, in patients with PA, sympathetic neural outflow is suppressed, and the sympathetic response to intravenous infusion of a high physiological dose of ANG II is augmented, and these responses are normalized with normalization of the renin-angiotensin system activity after unilateral adrenalectomy. These observations suggest that lowered renin-angiotensin system activity causes suppressed sympathetic nerve activity in patients with PA and that ANG II stimulates the sympathetic nervous system in humans.
Perspectives and Significance
In the present study, in PA patients, we confirmed that MSNA was suppressed and found that an infusion of a high physiological dose of ANG II increased MSNA in the patients. We have demonstrated a mechanism that accounts partly for the suppressed sympathetic nerve activity present in PA. In future studies, we would like to examine the effect of a smaller physiological dose of ANG II on MSNA. In addition, it will be of interest to determine the effects of mineralocorticoid receptor blockade on MSNA in PA patients and hypertensive control subjects, that is, with mineralocorticoid receptor blockade, plasma renin activity may increase while negating the effects of the aldosterone.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2008 the American Physiological Society