Aldosterone Excess or Escape: Treating Resistant Hypertension

Authors

  • Samira Ubaid-Girioli MD, PhD,

    1. From the Section of Cardiovascular Pharmacology and Hypertension, Department of Pharmacology,
    Search for more papers by this author
  • Leoní Adriana De Souza PharmD, PhD,

    1. From the Section of Cardiovascular Pharmacology and Hypertension, Department of Pharmacology,
    Search for more papers by this author
  • Juan Carlos Yugar-Toledo MD, PhD,

    1. From the Section of Cardiovascular Pharmacology and Hypertension, Department of Pharmacology,
    Search for more papers by this author
  • Luiz Cláudio Martins MD, MSc,

    1. From the Section of Cardiovascular Pharmacology and Hypertension, Department of Pharmacology,
    Search for more papers by this author
  • Sílvia Ferreira-Melo PharmD, PhD,

    1. From the Section of Cardiovascular Pharmacology and Hypertension, Department of Pharmacology,
    Search for more papers by this author
  • Otávio Rizzi Coelho MD, PhD,

    1. the Department of Internal Medicine, Cardiology Unit,
    Search for more papers by this author
  • Cristina Sierra MD, PhD,

    1. Faculty of Medical Sciences, State University of Campinas (UNICAMP), Campinas, SP, Brazil; the Hypertension Unit, Department of Internal Medicine, Hospital Clinic, School of Medicine, University of Barcelona, Barcelona, Spain;
    Search for more papers by this author
  • Antonio Coca MD, PhD,

    1. Faculty of Medical Sciences, State University of Campinas (UNICAMP), Campinas, SP, Brazil; the Hypertension Unit, Department of Internal Medicine, Hospital Clinic, School of Medicine, University of Barcelona, Barcelona, Spain;
    Search for more papers by this author
  • Eduardo Pimenta MD,

    1. the Endocrine Research Centre and Clinical Centre of Research Excellence in Cardiovascular Disease and Metabolic Disorders, University of Queensland School of Medicine, Princess Alexandra Hospital, Brisbane, Australia
    Search for more papers by this author
  • Heitor Moreno MD, PhD

    1. From the Section of Cardiovascular Pharmacology and Hypertension, Department of Pharmacology,
    Search for more papers by this author

Heitor Moreno, MD, PhD, Cardiovascular Pharmacology and Hypertension Division, Department of Pharmacology, Faculty of Medical Sciences, State University of Campinas (UNICAMP), PO Box 6111, 13083-970, Campinas, São Paulo, Brazil
E-mail: hmoreno@uol.com.br

Abstract

Aldosterone excess or “escape” can occur after treatment with medications that block the renin-angiotensin-aldosterone system or in undiagnosed primary aldosteronism. Spironolactone is thought to be an important addition to resistant hypertension (RH) treatment. In this study, resistant (RH) and controlled (CH) hypertensives and normotensive patients were submitted to echocardiography, flow-mediated vasodilation, carotid intima-media wall thickness studies, renin plasma activity, and aldosterone plasma levels and plasma and urinary sodium and potassium concentrations at baseline (pre-spironolactone phase). Subsequently, for only RH and CH groups, 25 mg/d spironolactone was added to preexisting treatments over 6 months. Afterwards, these parameters were reassessed (post-spironolactone phase). The RH and CH groups achieved reductions in blood pressure (P<.001), decreases in left ventricular hypertrophy (P<.001), improved diastolic function (Kappa index RH: 0.219 and Kappa index CH: 0.392) and increases in aldosterone concentrations (P<.05). The RH group attained improved endothelium-dependent (P<.001) and independent (P=.007) function. Optimized RH treatment with spironolactone reduces blood pressure and improves endothelial and diastolic function and left ventricular hypertrophy despite the presence of aldosterone excess or escape.

Until relatively recently, the mineralocorticoid hormone aldosterone was thought to be produced uniquely in the adrenal cortex and to act exclusively to promote sodium retention and potassium excretion. However, it is now known that aldosterone also acts on nonepithelial tissues, such as heart and blood vessels; high levels of this hormone can cause cardiovascular damage.1

There is an emerging understanding of the role of aldosterone as a cardiovascular risk factor. Some studies report that the “aldosterone escape phenomenon” with the use of angiotensin-converting enzyme inhibitors (ACEIs), angiotensin II receptor blockers (ARBs), and/or diuretics2,3 may be a cause of resistance to antihypertensive therapy. Also, primary aldosteronism has been implicated as a cause of resistance to treatment; 20% of patients with resistant hypertension (RH) may have primary aldosteronism. Many studies have shown the antihypertensive efficacy of aldosterone antagonists as an add-on agent in the treatment of RH.4–8 Interestingly, these medications cause a significant reduction in blood pressure (BP) independent of the aldosterone and renin levels or their ratio.5,9–12 The use of a mineralocorticoid receptor antagonist may also be indicated as an “organ-protecting” drug. However, until now the long-term effects of this class of antihypertensive medications on cardiovascular remodeling and function in RH patients have not been clarified.

This study investigated the effects of spironolactone on the BP, cardiovascular function and remodeling in optimally treated RH patients who had been taking a renin angiotensin system blocker (ACEIs and/or ARBs) plus a diuretic for at least 6 months.

Methods

Ethical Issues

This study was approved by the Institutional Research Ethics Committee of the State University of Campinas, São Paulo, Brazil. All of the participants were aware of the investigative nature of the study and provided informed written consent before participation.

Enrollment Procedures

Patients were allocated to resistant hypertensive (RH; n=39) or controlled hypertensive groups (CH; n=32). A control group consisted of 37 normotensive volunteers (CONT). All patients were nonsmokers. The volunteers underwent a physical examination, electrocardiography, and laboratory tests to exclude secondary forms of hypertension. Exclusion criteria included secondary forms of hypertension, impaired renal function, ischemic heart disease, liver disease, stroke, peripheral vascular disease, or other major diseases.

Study Protocol

The 71 hypertensive patients were evaluated for 6 months before the start of the study, and pharmacologic and nonpharmacologic therapies were optimized. Guidance was provided on the control of dietary salt which was confirmed by the measurement of urinary sodium.

Adherence

During this period a pharmacist provided the patients with prescribed medications. Adherence was measured by a pill count with patients who used ≥80% of each prescribed medicine being classified as adherent.13 Compliance to treatment was also assessed using the Morisky et al.14 test with only the patients who scored 4 points being considered adherent.15 The result of adherence before the start of the protocol was 85%.

Data Collection

Data collection was performed at 2 phases. At the start of the study (pre-spironolactone phase) all patients were evaluated with office BP, flow-mediated vasodilation (FMD), carotid intima-media wall thickness (IMT), plasma renin activity, plasma aldosterone concentration (PAC), and plasma and urinary sodium and potassium levels. After these exams, spironolactone 25 mg/d was added to the treatment of both RH and CH patients for 6 months; at the end of the study (post-spironolactone phase), all examinations were repeated. Normotensive patients received a placebo.

Blood Samples

Baseline blood samples for the quantification of plasma renin activity and PAC were collected at 8 am after overnight fasting during which time the volunteers rested in the supine position for 8 hours followed by 1 hour in an upright position in an air-conditioned room (22–24°C).

BP Measurements

Office BP Measurements.  The average of 2 readings was noted with hypertension being defined as a systolic BP (SBP) ≥140 mm Hg and/or a diastolic BP (DBP) ≥90 mm Hg on at least 3 different sets of measurements taken at 2-week intervals.16

Intermittent 24-Hour Ambulatory BP Monitoring.  Twenty-four hour ambulatory BP monitoring was carried out with an automatic oscillometric device. Patients engaged in their normal daily activities and BP was measured automatically at 20-minute intervals during the whole 24-hour period. The following parameters were measured: average 24-hour systolic, diastolic, mean, and pulse (difference between systolic and diastolic) pressures. This assessment was used only to exclude patients with white coat hypertension in the prestudy phase.17

FMD Technique.  Endothelial function was assessed by examining the brachial artery responses to endothelium-dependent (flow-mediated) and -independent (glyceryl-trinitrate [GTN]-mediated) stimuli18 using ACUSON CV 70 ultrasound equipment (Siemens Medical, Erlangen, Germany) and a high resolution linear vascular transducer 5–13 MHz.

Carotid IMT.  To measure the carotid IMT, high-resolution ultrasonography was performed on the left and right distal common carotid arteries.

Echocardiography.  The measurements of the dimensions of the left ventricle (LV) were performed using the American Society of Echocardiography (ASE) recommendations19 utilizing 2-dimensional targeted M-mode echocardiography. Diastolic and systolic LV diameters and the interventricular septal wall thickness and posterior wall thickness at the end of diastole were measured according to the QRS wave of electrocardiography. LV mass was calculated by the ASE recommended formula.20 Diastolic dysfunction was assessed by mitral valve Doppler and tissue Doppler examinations and was classified as grades I to IV according to published criteria.21 Normal diastolic function was defined when the E wave of the Doppler mitral flow was higher than the A wave and when the E′ wave of the tissue Doppler was higher then the A′ wave. In grade I diastolic dysfunction, the E wave was smaller than the A wave and the E′ wave was smaller than A′ wave. Grade II diastolic dysfunction was defined when the E wave was greater than the A wave and the E′ wave was smaller than the A′ wave. For grade III diastolic dysfunction, the E wave had to be greater than the A wave (generally 2 times higher) and the E′ wave smaller than the A′ wave but reversible using the Valsalva maneuver. And finally, grade IV diastolic dysfunction presented with the same pattern as grade III but was not reversible using the Valsalva maneuver.

Statistical Analysis

The classifying variables were evaluated using the chi-squared test or the likelihood ratio test. Quantitative variables are expressed as means, standard deviation, median, and minimum and maximum values. The variables were evaluated by parametric and nonparametric tests. The Pearson correlation coefficient and multiple linear regression analysis were used when appropriate. P values <.05 were considered statistically significant. The Kappa index was used for measuring agreement with categorical data (pre- and post-spironolactone phases).22

Results

Clinical Characteristics

The characteristics of the participants are shown in Table I. In the pre-spironolactone phase, the SBP and DBP of the 3 groups were significantly different (P<.001).

Table I.   General Characteristics of the Hypertensive and Normotensive Groups
 RH,n=39CH,n=32CONT,n=37
  1. Abbreviations: CH, controlled hypertension; CONT, control; HDLC, high-density lipoprotein cholesterol; LDLC, low-density lipoprotein cholesterol; RH, resistant hypertension. RH vs CONT, aP<.05; RH vs CH, bP=.0475; RH and CH vs CONT; cP<.05.

Glycemia (mg/dL)100.0±16.9a93.9±13.1103.6±5.4
Cholesterol (mg/dL)204.2±48.4c201.3±34.2182.9±32.2
HDLC (mg/dL)49.7±16.150.1±14.251.1±8.1
LDLC (mg/dL)132.3±45.6c25.8±30.3106.6±24.7
Triglycerides (mg/dL)158.1±96.2153.9±76.6122.8±80.2
Urea (mg/dL)32.8±13.632.4±11.328.2±7.2
Creatinine (mg/dL)1±0.3b0.8±0.20.8±0.2
Uric acid (mg/dL)6.1±1.7c5.6±1.24.8±1.0
Sodium (mEq/L)140.5±2.6140.9±1.7140.5±1.2
Potassium (mEq/L)4.2±0.44.1±0.34.2±0.3
Urinary sodium (mEq/24 h)218.2±117.9192.9±76.5159.6±42.4
Urinary potassium (mEq/24 h)47.0±21.946.2±17.744.5±13.1

RH patients received a mean of 3.6 medications daily. During optimization of the therapeutic regime, the most frequently prescribed medications were diuretics (94.8%), calcium channel blockers (84.6%), ARBs (79.5%), ACEIs (25.6%), and β-blockers (53.8%). CH patients took a mean of 2.5 medications daily with the prescribed medications being diuretics (93.7%), channel calcium blockers (28.1%), ARBs (31.2%), ACEIs (40.6%), and β-blockers (46.9%). The same medications were maintained with the addition of 25 mg/d spironolactone. In the RH and CH groups, the office BP was reduced significantly after 6 months of treatment with spironolactone (pre-SBP vs post-SBP; P<0.001 and pre-DBP vs post-DBP; P<0.01) as is illustrated in Figure 1 and Figure 2. A total of 33% of patients in the RH group and 28% in the CH group were able to reduce the dosage of drugs after the use of spironolactone, but without decreasing the number of drugs.

Figure 1.

Systolic blood pressure (SBP; mm Hg) in normotensive (•—control [CONT]) and hypertensive (○—controlled hypertension [CH] and △—resistant hypertension [RH]) groups, in pre- and post-spironolactone phases. The values are expressed as mean ± standard deviation. Pre-spironolactone—RH vs CONT, RH vs CH, CH vs CONT; *P<.001. Post-spironolactone—RH vs CONT, RH vs CH; †P<.001. SBP pre × SBP post; ‡P<.001.

Figure 2.

Diastolic blood pressure (DBP; mm Hg) in normotensive (•—control [CONT]) and hypertensive (○—controlled hypertension [CH] and △—resistant hypertension [RH]) groups, in pre- and post-spironolactone phases. The values are expressed as mean ± standard deviation. Pre-spironolactone—RH vs CONT, RH vs CH, CH vs CONT; *P<.001. Post-spironolactone—RH vs CONT, RH vs CH, CH vs CONT; †P<.001. DBP pre vs DBP post; ‡P<.001.

The most common adverse effects, which occurred in 6% of patients, were nausea (6%), gynecomastia (5%), and breast tenderness (5%). After introducing spironolactone, dyspnea and fatigue improved in 32% of the RH patients and palpitations in 23%, with CH patients improving in 20% and 11%, respectively.

Optimal antihypertensive treatment alone was insufficient to improve endothelial function in RH. The addition of an aldosterone antagonist for 6 months improved endothelium-dependent and -independent vasodilation (FMD; P<0.001 and nitroglycerine; P=0.007) as shown in Figure 3. CH patients had normal endothelium-dependent and -independent vasodilation before the addition of spironolactone; this did not differ at the end of the study.

Figure 3.

Dependent (flow-mediated vasodilation [FMD]) and independent (nitroglycerine [NTG]) endothelium vasodilation pre- and post-association with spironolactone. The values are expressed as median/minimum/maximum. Pre-phase FMD—RH vs CONT; *P<.05. Pre-phase NTG—RH vs CONT, RH vs CH; *P<.05. Post-phase NTG—RH vs CONT; †P<.05. FMD RH pre vs FMD RH post; ‡P<.001. NTG RH pre vs NTG RH post; ‡P=.007. CH indicates controlled hypertension; CONT, control; RH, resistant hypertension; CH = controlled hypertension.

The LV mass index in the pre-spironolactone phase was statistically different among the 3 groups. The addition of spironolactone produced a significant reduction in the LV mass index in RH and CH patients calculated using both the body surface (bs) and height2.7 (LV mass/bs; P=0.001 and LV mass/h2.7; P=0.05). At the beginning of the study, 95% of RH and 71.8% of CH patients had diastolic dysfunction. After the introduction of spironolactone, the RH and CH patients presented with improvements of 7.7% and 15.6% in diastolic dysfunction, respectively (Kappa index RH: 0.219 and Kappa index CH: 0.392).

Treatment with other drugs before the addition of spironolactone proved to be effective to normalize RH vascular remodeling, as patients began the specific study with normal IMT patterns.

In the pre-spironolactone phase, there were no statistically significant differences among the groups with respect to the biochemical markers. However, there was a significant increase in plasma aldosterone levels in RH and CH patients and an increase in plasma potassium in RH patients (P=0.004) 6 months after taking a mineralocorticoid antagonist (P<0.005) as shown in Table II.

Table II.   Biochemical Markers of Hypertensive and Normotensive Groups, in Pre- and Post-Spironolactone Association
 RH Pre-SpiroRH Post-SpiroCH Pre-SpiroCH Post-SpiroCONT Pre-SpiroCONT Post-Spiro
  1. Abbreviations: ARR, aldosterone/renin ratio; CH, controlled hypertension; CONT, control; K+URI, urinary potassium; Na+URI, urinary sodium; PAC, plasma aldosterone concentration; PRA, plasma renin activity; post-spiro, post-spironolactone; pre-spiro, pre-spironolactone; RH, resistant hypertension. Post-phase—RH vs CONT, *P<.05. CH vs CONT and RH, P<.05. Pre-phase vs post-phase, P<.005.

PRA (ng/mL/h)5.1±7.15.0±9.4 4.6±8.14.7±3.7†1.7±0.92.1±1.5
PAC (ng/mL)15.4±8.219.2±9.0‡16.8±5.422.3±12.6‡15.4±6.915.2±6.7
Sodium (mEq/L) 140.5±2.6139.9±3.4140.9±1.7140.4±2.0140.5±1.2140.5±1.1
Potassium (mEq/L) 4.2±0.44.4±0.5*‡4.1±0.34.3±0.4 4.2±0.34.2±0.2
Na+URI (mEq/24 h)218.2±117.9239.0±126.6*192.9±76.5200.6±81.6159.6±42.4155.9±44.2
K+URI (mEq/24 h)47.0±21.948.9±16.946.2±17.747.4±17.544.5±13.142.7±12.7
ARR13.2±15.315.0±22.48.6±6.510.3±21.010.7±5.68.6±3.6

Discussion

The patients in this study were included after optimization of therapy, following the guidelines of the Seventh Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure,23 with a compliance of 85%. Patients with RH had evidence of target organ involvement such as LV hypertrophy, endothelial dysfunction, and diastolic dysfunction. There is increasing evidence that aldosterone exerts major adverse cardiovascular effects through classical mineralocorticoid receptors (MR) in nonepithelial tissues such as the heart and blood vessels. This nonepithelial role of aldosterone has been underscored by the recent Randomized Aldactone Evaluation Study (RALES) and the Eplerenone Post-AMI Heart Failure Efficacy and Survival Study (EPHESUS). These studies showed that further benefits could be obtained in patients receiving an ACEI or ARB. With the addition of an MR antagonist, the long-term effect of aldosterone was not inhibited in some subjects with the addition of an ACEI or an ARB, so the possibility of organ damage due to so-called “breakthrough” aldosterone cannot be ignored. Nonepithelial MR-mediated effects played a major role in this aldosterone effect.24 Thus, the benefits of spironolactone stated in our study may be explained by the mechanisms described above.

The RH and CH groups maintained the same number of drug classes after 6 months of treatment with 25 mg/d spironolactone. Other authors, however, have reported a reduction in the number of classes of antihypertensives after the use of spironolactone.25 The doses of medication, however, were reduced by 33% in the RH group and by 28% in the CH group.

Compared to the CH and CONT groups, this study demonstrates that before the addition of spironolactone, BP control, carotid IMT, diastolic dysfunction, LV hypertrophy, and FMD were worse in the RH group. Other publications have reported an association between endothelial dysfunction and LV diastolic function26 and between endothelial dysfunction and LV hypertrophy.27 There now is extensive experimental and growing clinical evidence for an important nonphysiological role for aldosterone in the pathology of cardiac and renal disease. Classical effects of aldosterone are mediated via its nuclear receptor. These effects of aldosterone have been demonstrated in the kidney, vascular smooth muscle cell, and leukocytes. It has been hypothesized that cardiac damage induced by aldosterone is independent of the presence of hypertension.28 Certainly, both effects of aldosterone can be responsible for the cardiovascular damage we found in the RH group.

Decreases in BP of 21.6 mm Hg for SBP and 9.1 mm Hg for DBP were noted for the RH and CH groups with the addition of spironolactone to antihypertensive treatment. Reductions in BP have been reported in other studies. Lane et al. reported mean drops in SBP of 21.7 mm Hg and DBP of 8.5 mm Hg.4 Similar reductions were observed by Chapman et al.5 (21.9 mm Hg and 9.5 mm Hg, respectively) and by Pimenta et al.7 (25 mm Hg and 12 mm Hg, respectively).

This study found evidence of vascular dysfunction in the pre-spironolactone phase characterized as a reduction in both endothelial-dependent and -independent vasodilation of RH patients. On the other hand, endothelial-dependent vasodilation was normal in both CONT and CH groups. Both elevated BP and aldosterone excess can impair endothelial function.26,27,29 Aldosterone can directly and indirectly influence vasoconstrictor mechanisms and the effects of endothelial-dependent dilators via angiotensin II. Angiotensin II induces endothelial dysfunction as a result of increased oxidative stress which may lead to a reduction in the bioavailability of nitric oxide (NO). However, mineralocorticoid receptor antagonism by spironolactone improves the vascular relaxation. Pu et al.30 demonstrated that the infusion of aldosterone in rats diminishes endothelial-dependent relaxation and is associated with oxidative stress of the vascular wall. This effect was reversed by the mineralocorticoid receptor antagonist spironolactone.31 In humans endothelial-dependent relaxation is also improved by these medications with improved bioavailability of NO, the mechanism most responsible.32

Elevated plasma aldosterone and renin levels are associated with LV hypertrophy.33 In our study, the addition of spironolactone to preexistent therapy reduced the LV mass index in both RH and CH groups. This probably contributes to an improvement in cardiovascular risk.34 This reinforces the findings of Taniguchi, et al. who demonstrated that 25 mg/d spironolactone given with an ARB for 6 months reduced hypertrophy despite the smaller reduction in BP compared to ARB treatment alone. These results indicate that the regression of concentric LV hypertrophy due to spironolactone treatment might be independent of reductions in BP.35

Diastolic dysfunction was improved in the RH and CH groups in 7.7% and 15.6%, respectively after the use of spironolactone. This finding was associated with an improvement in dyspnea, fatigue, and palpitations reported by the patients and also demonstrated by Sato et al.34

It is known that vascular remodeling, detected by an evaluation of carotid IMT, is associated with higher risks of cardiovascular, coronary artery, and encephalic events.36–38 Interestingly, we found that carotid IMT was already within the normal range before the addition of spironolactone; this might reflect the effects of optimized treatment. Curiously, the IMT in RH patients was greater than in the CH group. This finding can be explained by the higher BP levels in RH group.

Although patients presented high potassium plasma levels after the use of spironolactone, none of the patients had potassium levels higher than 6 mEq/dL or electrocardiographic alterations) which would justify the interruption of the treatment. Probably, the low dose of spironolactone used here was responsible for this result.

The biochemical measurements of renin-angiotensin-aldosterone system markers did not present significant differences among the groups in the pre-spironolactone phase. However, after spironolactone, PAC increased in both RH and CH groups. It is known that the competitive antagonism of spironolactone increases the circulating levels of aldosterone.39 For this reason, the high levels of aldosterone we found do not characterize aldosterone escape. Since cardiovascular risk factors (high SBP and DBP, LV hypertrophy, endothelial dysfunction, and diastolic dysfunction) improved with the use of spironolactone, we suggest that the activated mineralocorticoid receptors which are blocked by this agent may be responsible for cardiac remodeling and endothelial and diastolic dysfunction, with increases in BP. The addition of spironolactone to antihypertensive treatment, with or without aldosterone excess or escape, significantly improves BP levels, left ventricular hypertrophy, diastolic function and endothelial function, and is well tolerated.

Conclusion

As far as we know, this is the first study in which the long-term effects of aldosterone antagonists on BP levels, left ventricular hypertrophy, diastolic function, and endothelial function were observed simultaneously. These data suggest that spironolactone or another aldosterone antagonist should probably be used more widely in the treatment of RH. Obviously, further investigations are necessary in larger trials.

Also, we demonstrated that this mineralocorticoid antagonist can be used in RH independently of plasma aldosterone levels. This is useful for those clinicians who do not have to investigate hormone plasma levels.

Disclosures:  This study was supported by CAPES (Coordenação de Aperfeiçoamento de Pessoal de Nível Superior), CNPq (Conselho Nacional de Desenvolvimento Científico e Tecnologia), and FAPESP (Fundação de Amparo à Pesquisa do Estado de São Paulo), Brazil. Spironolactone was provided by Biolab (São Paulo, Brazil).

Ancillary