- Email: [email protected]
Spironolactone inhibits the transcardiac extraction of aldosterone in patients with congestive heart failure
Journal of the American College of Cardiology © 2000 by the American College of Cardiology Published by Elsevier Science Inc.
Vol. 36, No. 3, 2000 ISSN 0735-1097/00/$20.00 PII S0735-1097(00)00796-8
Spironolactone Inhibits the Transcardiac Extraction of Aldosterone in Patients With Congestive Heart Failure Takayoshi Tsutamoto, MD, Atsuyuki Wada, MD, Keiko Maeda, MD, Naoko Mabuchi, MD, Masaru Hayashi, MD, Takashi Tsutsui, MD, Masato Ohnishi, MD, Masahide Sawaki, MD, Masanori Fujii, MD, Takehiro Matsumoto, MD, Hajime Horie, MD, Yoshihisa Sugimoto, MD, Masahiko Kinoshita, MD Otsu, Japan The study evaluated the transcardiac extraction or spillover of aldosterone (ALDO) in normal subjects and in patients with congestive heart failure (CHF). BACKGROUND Aldosterone promotes collagen synthesis and structural remodeling of target organs such as the heart. Spironolactone, an ALDO receptor antagonist, has recently been reported to reduce the mortality of patients with CHF; however, the effects of spironolactone on the transcardiac gradient of ALDO have not been clarified. METHODS We measured plasma ALDO in the aortic root (AO) and coronary sinus (CS) in normal subjects and 113 consecutive CHF patients and also measured plasma procollagen type III aminoterminal peptide (PIIINP) in CS, a biochemical marker of myocardial fibrosis. RESULTS Plasma ALDO was significantly lower in the CS than in the AO in normal subjects (n ⫽ 15; 61.2 ⫾ 9.3 vs. 83.1 ⫾ 11.8 pg/ml, p ⬍ 0.0001). In 96 CHF patients who did not receive spironolactone, plasma ALDO was significantly lower in the CS than in the AO (59.3 ⫾ 3.9 vs. 73.8 ⫾ 4.9 pg/ml, p ⬍ 0.0001). In contrast to the difference in these 96 patients, there was no significant difference in ALDO between the AO and CS in 17 patients who received spironolactone (127.4 ⫾ 20 vs. 124.0 ⫾ 19 pg/ml, p ⫽ 0.50). Stepwise multivariate analyses showed that spironolactone therapy had an independent and significant negative relationship with the transcardiac gradient of plasma ALDO in patients with CHF. In addition, significant positive correlations were seen between the transcardiac gradient of plasma ALDO and PIIINP (r ⫽ 0.565, p ⬍ 0.0001) and the left ventricular end-diastolic volume index (r ⫽ 0.484, p ⬍ 0.0001). CONCLUSIONS These results indicate that plasma ALDO is extracted through the heart in normal subjects and in CHF patients who do not receive spironolactone and that spironolactone inhibits the transcardiac extraction of ALDO in CHF patients, suggesting that spironolactone blocks the effects of ALDO on the failing heart in patients with CHF. (J Am Coll Cardiol 2000;36: 838 – 44) © 2000 by the American College of Cardiology OBJECTIVES
Despite the outstanding success of angiotensin-converting enzyme (ACE) inhibitors in the treatment of congestive heart failure (CHF), the mortality rate of CHF patients remains higher than that of normal subjects. This high mortality rate is partly due to the aldosterone escape phenomenon in patients who received ACE inhibitors (1– 4). Moreover, the plasma angiotensin II (Ang II) level is not constantly decreased during long-term treatment with ACE inhibitors (5), partly because Ang II may be produced by alternative pathways (6). Therefore, angiotensin type 1 receptor antagonists or aldosterone (ALDO) receptor antagonists such as spironolactone may have additional advantages over ACE inhibitors (3,7). Recently, spironolactone was shown to reduce the mortality of CHF patients who had already received ACE inhibitors, digitalis, and diuretics. In the Randomized Aldactone Evaluation Study (RALES) (3,8), both sudden death and CHF death were significantly reduced with spironolactone. However, despite the great impact of the RALES results, the mechanism of the From the First Department of Internal Medicine, Shiga University of Medical Science, Tsukinowa, Seta, Otsu, Japan. Supported in part by Grants-in-Aid for Scientific Research of Japan. Manuscript received December 8, 1999; revised manuscript received March 15, 2000, accepted April 26, 2000.
improvement of mortality with spironolactone remains to be fully elucidated. Aldosterone displays both myocardial and renal effects that may have profound implications for left ventricular remodeling (9). Significant associations were found between plasma ALDO and left ventricular mass in patients with hypertension as well as in a population-based sample, suggesting that aldosterone may be an important factor in the modulation of cardiac structure (10 –12). Recently, it has been reported that mineralocorticoid receptor, which mediates the action of ALDO, is expressed in cardiomyocytes, endothelial cells, and fibroblasts in the human heart (13–15). Aldosterone, secreted from the adrenal gland, has been shown to stimulate cardiac collagen synthesis and fibroblast proliferation via activation of local mineralocorticoid receptors or via an unexplained indirect mechanism (16 –19). However, the relationship between the cardiac effect of ALDO and the mechanism of cardiac fibrosis by ALDO has not been fully elucidated. More recently, cardiac production of ALDO has been reported in rats (20), suggesting the possibility of cardiac production of ALDO in humans. However, a transcardiac gradient of ALDO in humans has not been demonstrated. In the present study, we measured plasma levels of ALDO both in the aortic root and the coronary sinus, and
Tsutamoto et al. Aldosterone Extraction Through the Failing Heart
JACC Vol. 36, No. 3, 2000 September 2000:838–44
Abbreviations and Acronyms ACE ⫽ angiotensin-converting enzyme ALDO ⫽ aldosterone Ang II ⫽ angiotensin II ANP ⫽ atrial natriuretic peptide AO ⫽ aortic root BNP ⫽ brain natriuretic peptide CHF ⫽ congestive heart failure CS ⫽ coronary sinus LVEDVI ⫽ left ventricular end-diastolic volume index LVEF ⫽ left ventricular ejection fraction PIIINP ⫽ procollagen type III aminoterminal peptide
we evaluated whether ALDO is extracted or spilled over through the heart in normal subjects and in patients with CHF. Moreover, we also evaluated the effects of spironolactone on the transcardiac gradient of ALDO and determined the relationship between the transcardiac gradient of ALDO and the left ventricular remodeling.
METHODS Patients. The subjects were 113 consecutive patients with symptomatic left ventricular dysfunction (left ventricular ejection fraction [LVEF] ⬍45%) without renal failure who underwent cardiac catheterization for clinical indications. Patients with angina pectoris, secondary hypertensive heart disease, renal failure, or liver dysfunction were excluded. Patients with coronary stenosis (⬎70%) were excluded. We also selected 15 age-matched normal subjects (age, 34 to 71; mean: 54 years) who were admitted complaining of chest pain and whose hearts proved to be normal by coronary angiography. Informed consent was obtained from all patients before participation in the study, and the protocol was approved by the Human Investigations Committee of our institution. The subjects were 75 men and 38 women ranging in age from 17 to 79 years (mean: 58 years); 68 patients had suffered a myocardial infarction more than three months before the study, 38 had dilated cardiomyopathy, 5 had hypertensive heart disease, and 2 had valvular heart disease. Sixty-eight patients with old myocardial infarction had previously received percutaneous transluminal coronary angioplasty (PTCA) in our hospital. Eighty-four patients were classified according to the standards of the New York Heart Association (NYHA) as functional class II, 25 patients as class III, and 4 patients as
839
class IV. At entry to the study, 76 patients were treated with furosemide, 17 with spironolactone (a dose of 25 to 50 mg daily), 79 with ACE inhibitors, 58 with digitalis, 37 with vasodilators, and 27 with beta-blockers. Most drugs had been administered for more than two months. Study protocol. All patients were premedicated with an oral dose of diazepam (5 mg) and rested in bed in the supine position for at least 20 min. Right-sided cardiac catheterization was performed using a 7F Swan-Ganz catheter. The heart rate was monitored by electrocardiography. Blood samples for measuring plasma ALDO were collected simultaneously from the aortic root (AO) and coronary sinus (CS). A 6F catheter (Goodman, Tokyo, Japan) for blood sampling was positioned in the CS, and the position of the catheter was confirmed by injection of contrast medium just after blood sampling. The validity of CS blood sampling was also verified by simultaneous measurements of the atrial natriuretic peptide (ANP) and brain natriuretic peptide (BNP) levels in the AO and CS. In 79 of 113 patients, blood samples for measuring the plasma levels of procollagen type III aminoterminal peptide (PIIINP) were drawn from the CS. A Swan-Ganz catheter was inserted through the right femoral vein into the main pulmonary artery, where the pressure was measured. The catheter was then advanced into the pulmonary artery, and the pulmonary capillary wedge pressure was measured by inflating the balloon. Cardiac output was determined by the thermodilution method immediately after blood collection. Left ventriculography was performed using contrast medium after the hemodynamic measurements and blood sampling. Measurement of neurohumoral factors. Blood for the measurement of the plasma levels of ANP and BNP was transferred to a chilled tube containing EDTA (1 mg/ml) and aprotinin (500 kallikrein inactivator units/ml), and then centrifuged at 3,000 rpm for 15 min at 4°C. The plasma thus obtained was stored at ⫺30°C until assayed. Plasma concentrations of ANP and BNP were measured with a specific immunoradiometric assay for using a commercial kit (Shionogi, Osaka, Japan) as previously reported (21). Blood for measurement of the plasma levels of ALDO was transferred to a chilled tube containing EDTA (1 mg/ml), centrifuged at 3,000 rpm for 15 min at 4°C, and the plasma thus obtained was stored at ⫺30°C until it was assayed. Plasma ALDO levels were measured using a commercial radioimmunoassay kit. This assay system for
Table 1. Hemodynamic and Neurohumoral Data
Mild CHF (n ⫽ 84) (SE) Severe CHF (n ⫽ 29) (SE)
LVEF (%)
MPAP (mm Hg)
PCWP (mm Hg)
LVEDP (mm Hg)
LVEDVI (l/min/m2)
ANP (pg/ml)
BNP (pg/ml)
ALDO in AO (pg/ml)
ALDO in CS (pg/ml)
35.6 (0.7) 22.9* (1.8)
16.9 (0.4) 30.6* (2.1)
8.7 (0.3) 20.7* (1.8)
11.3 (0.5) 25.4* (1.9)
122 (3.9) 196* (12)
71.7 (11.3) 207* (24)
87.5 (8.3) 459* (72)
73.6 (5.6) 119.6* (15)
61.2 (4.4) 102* (14)
LVEF ⫽ left ventricular ejection fraction; MPAP ⫽ mean pulmonary arterial pressure; PCWP ⫽ pulmonary capillary wedge pressure; LVEDP ⫽ left ventricular end-diastolic pressure; LVEDVI ⫽ left ventricular end-diastolic volume index; ANP ⫽ atrial natriuretic peptide; BNP ⫽ brain natriuretic peptide; ALDO ⫽ aldosterone; AO ⫽ ascending aorta; CS ⫽ coronary sinus; SE ⫽ standard error. *p ⬍ 0.0001 vs. mild CHF.
840
Tsutamoto et al. Aldosterone Extraction Through the Failing Heart
JACC Vol. 36, No. 3, 2000 September 2000:838–44
Table 2. Univariate and Multivariate Linear Model of Left Ventricular End-Diastolic Volume Index in 113 Patients With CHF
Variable Age (yr) Gender (male ⫽ 1; female ⫽ 0) Etiology (ICM ⫽ 1; non-ICM ⫽ 0) NYHA functional class LVEF (%) Mean blood pressure (mm Hg) Cardiac index (l/min/m2) Left ventricular end-diastolic pressure (mm Hg) Atrial natriuretic peptide (pg/ml) Brain natriuretic peptide (pg/ml) ALDO in AO (pg/ml) (AO-CS) ALDO (pg/ml) Treatment (treatment ⫽ 1; no treatment ⫽ 0) Spironolactone Diuretics Digitalis ACE inhibitors Vasodilators Beta-blockers
Univariate Correlation Coefficient
p Value
⫺0.115 ⫺0.223 ⫺0.536 0.473 ⫺0.665 ⫺0.236 ⫺0.029 0.337 0.487 0.351 0.317 0.484
0.283 0.0057 ⬍0.0001 ⬍0.0001 ⬍0.0001 0.003 0.758 0.0003 ⬍0.0001 0.0002 0.0006 ⬍0.0001
0.302 0.473 0.419 ⫺0.174 ⫺0.290 0.117
0.005 ⬍0.0001 ⬍0.0001 0.081 0.006 0.128
Multivariate Beta Coefficient
⫺3.4
0.720
p Value
⬍0.0001
⬍0.0001
ICM ⫽ ischemic cardiomyopathy; LVEF ⫽ left ventricular ejection fraction; ALDO ⫽ plasma aldosterone; AO ⫽ aortic root; CS ⫽ coronary sinus; ACE ⫽ angiotensin-converting enzyme.
plasma ALDO did not cross-react with angiotensin I or II, spironolactone, ANP, or BNP. Plasma levels of PIIINP were measured with a specific immunoradiometric assay using a commercial kit (CIS Bio International, Nagoya, Japan).
Statistical analysis. All results are expressed as the mean ⫾ SEM. Univariate analyses were performed using the Student t test. To evaluate the contribution of ALDO to left ventricular remodeling, univariate and stepwise multivariate analyses were used among the 18 variables. Univariate and
Table 3. Univariate and Multivariate Linear Model of Transcardiac Gradient of Plasma Aldosterone in 113 Patients With CHF
Variable Age (yrs) Gender (male ⫽ 1; female ⫽ 0) Etiology (ICM ⫽ 1, non-ICM ⫽ 0) NYHA functional class LVEF (%) Mean blood pressure (mm Hg) Cardiac index (l/min/m2) Left ventricular end-diastolic blood pressure (mm Hg) Left ventricular end-diastolic volume index (ml/m2) Atrial natriuretic peptide (pg/ml) Brain natriuretic peptide (pg/ml) ALDO (pg/ml) Treatment (treatment ⫽ 1; no treatment ⫽ 0) Spironolactone Diuretics Digitalis ACE inhibitors Vasodilators Beta-blockers
Univariate Correlation Coefficient
p Value
⫺0.031 ⫺0.067 ⫺0.109 0.280 ⫺0.224 ⫺0.180 ⫺0.097 0.250
0.744 0.438 0.251 0.0029 0.018 0.059 0.309 0.008
0.484
⬍0.0001
0.217
0.370 0.231 0.371
⬍0.0001 0.389 ⬍0.0001
0.10
⫺0.207 0.114 0.056 ⫺0.176 ⫺0.089 ⫺0.062
0.027 0.228 0.554 0.062 0.348 0.512
Multivariate Beta Coefficient
⫺14.7
p Value
⬍0.0001
0.0052
0.0115
ICM ⫽ ischemic cardiomyopathy; LVEF ⫽ left ventricular ejection fraction; ALDO ⫽ plasma aldosterone; AO ⫽ aortic root; ACE ⫽ angiotensin-converting enzyme.
JACC Vol. 36, No. 3, 2000 September 2000:838–44
Figure 1. Plasma aldosterone (ALDO) concentrations in the aortic root (AO) and coronary sinus (CS) in normal subjects.
stepwise multivariate linear regression analyses were also used to detect independent predictors of the transcardiac gradient of plasma ALDO among the 18 variables. Linear regression analysis was used to determine the relationship between continuous variables. A p value ⬍0.05 was regarded as significant.
RESULTS Difference between plasma concentration of ALDO in aortic root and coronary sinus. In the 15 age-matched normal subjects, the plasma ALDO level was significantly lower in the CS than in the AO (Fig. 1) (61.2 ⫾ 9.3 vs. 83.1 ⫾ 11.8 pg/ml, p ⬍ 0.0001). In 113 consecutive CHF patients, plasma ALDO was also significantly lower in the CS than in the AO (69.0 ⫾ 4.8 vs. 81.9 ⫾ 5.4 pg/ml, p ⬍ 0.0001). In the 96 patients without spironolactone treatment, plasma ALDO was significantly lower in the CS than in the AO (Fig. 2) (59.3 ⫾ 3.9 vs. 73.8 ⫾ 4.9 pg/ml, p ⬍ 0.0001). In contrast to the difference in the patients who did not receive spironolactone, there was no significant difference of ALDO between the AO and CS in the 17 patients who received spironolactone (Fig. 2) (127.4 ⫾ 20 vs. 124.0 ⫾ 19 pg/ml, p ⫽ 0.50). Plasma ALDO level was significantly higher in the patients who received spironolactone than in the patients who did not receive spironolactone. Hemodynamic and neurohumoral data. Patients were divided into two groups according to symptoms and were analyzed for hemodynamic and neurohumoral data (Table 1). Hemodynamic parameters such as pulmonary capillary wedge pressure, left ventricular end-diastolic pressure, and left ventricular end-diastolic volume index (LVEDVI) were significantly higher in severe CHF patients (NYHA class III or class IV) than in mild CHF patients (NYHA class II). The LVEF was significantly lower in severe CHF than in mild CHF patients. Neurohumoral factors such as ANP, BNP, and ALDO were significantly higher in severe CHF patients than in mild CHF patients. Plasma ALDO was
Tsutamoto et al. Aldosterone Extraction Through the Failing Heart
841
Figure 2. Plasma aldosterone (ALDO) concentrations in the aortic root (AO) and coronary sinus (CS) in patients with congestive heart failure (CHF). Left panel shows data of ALDO in 96 CHF patients who did not receive spironolactone; right panel shows data of ALDO in 17 CHF patients who received spironolactone. *, p ⬍ 0.0001 vs. the value of AO.
significantly lower in the CS than in the AO in mild CHF patients and in severe CHF patients. Relationship between plasma ALDO extraction across the heart and LVEDVI. Table 2 shows the results of univariate and multivariate analyses to assess the factors regulating the LVEDVI in 113 patients with CHF. According to stepwise multivariate analyses, only a high transcardiac gradient of plasma ALDO (p ⬍ 0.0001) and LVEF (p ⬍ 0.0001) were significant independent predictors of a large LVEDVI. Figure 3 shows the correlation between the LVEDVI and the plasma levels of ALDO in the AO (r ⫽ 0.317; p ⫽ 0.0006) and the transcardiac gradient of plasma ALDO (r ⫽ 0.484; p ⬍ 0.0001) in patients with CHF. A significant positive correlation existed between the transcardiac extraction of plasma ALDO and plasma level of PIIINP in the CS (Fig. 4; r ⫽ 0.565; p ⬍ 0.0001), and also a significant correlation between the plasma level of PIIINP in the CS and LVEDVI (r ⫽ 0.448; p ⬍ 0.0001). Relationship between plasma ALDO in aortic root and ALDO extraction across the heart. Table 3 shows the results of univariate and multivariate analyses among 18 variables to assess the factors regulating the transcardiac extraction of plasma ALDO. According to stepwise multivariate analyses, a high level of plasma ALDO in the AO and treatment with spironolactone were significant independent predictors of transcardiac extraction of plasma ALDO in patients with CHF. Figure 5 shows a significant correlation between the plasma level of ALDO in the AO and the transcardiac extraction of plasma ALDO by linear regression analysis in the 96 patients not treated with spironolactone (r ⫽ 0.759; p ⬍ 0.0001). There was no correlation between the plasma level of ALDO in the AO and the transcardiac extraction of plasma ALDO in 17 patients who received spironolactone, and most of the 17 patients who received spironolactone were on the right side of the linear regression line of the 96 patients who did not receive spironolactone (Fig. 5).
842
Tsutamoto et al. Aldosterone Extraction Through the Failing Heart
DISCUSSION It has been recently reported that mineralocorticoid receptors, which mediate the action of ALDO, are expressed in the human heart (13–15). Cardiac fibrosis in rats induced by ALDO infusion was not inhibited by concomitant administration of an ACE inhibitor or an Ang II receptor antagonist, losartan, but was completely blocked by concomitant administration of the mineralocorticoid receptor antagonist potassium canrenoate (22). However, the relationship between the cardiac effect of ALDO and the mechanism of cardiac fibrosis by ALDO has not been fully elucidated in vivo. We demonstrated here for the first time that plasma ALDO is extracted through the heart in normal subjects and in patients with CHF, and that a positive correlation exists between the transcardiac gradient of ALDO and LVEDVI. Moreover, spironolactone therapy inhibited the ALDO extraction through the failing heart. These findings indicate that the heart is a target organ of ALDO, and that spironolactone, a mineralocorticoid receptor antagonist, inhibits the ALDO extraction through the heart in patients with CHF. We also found a positive correlation between the transcardiac ALDO extraction and the plasma level of PIIINP, a possible biochemical marker of myocardial fibrosis (23,24) in the CS, suggesting that the increase of the transcardiac ALDO extraction in the failing heart stimulates myocardial collagen turnover, as shown in vitro (19). Our finding of a positive correlation between the transcardiac ALDO extraction and the plasma level of PIIINP also suggests that the sustained transcardiac ALDO extraction is an important modulator of left ventricular remodeling in the failing heart regardless of the plasma ALDO levels. Klappacher et al. (23) reported a significant positive correlation between serum PIIINP and the amount of myocardial collagen type III on cardiac biopsy in patients with CHF. Moreover, Host et al. (24) showed that after a myocardial infarction, the plasma level of PIIINP was higher in those patients with a poor prognosis. Therefore, the plasma level of PIIINP may be a biochemical marker of myocardial fibrosis and/or left
JACC Vol. 36, No. 3, 2000 September 2000:838–44
ventricular remodeling in patients with CHF. Indeed, in the present study, there was a positive correlation between the plasma level of PIIINP and LVEDVI. In addition, the positive correlation between the transcardiac ALDO extraction and the plasma level of PIIINP suggests an interaction between ALDO extraction and left ventricular remodeling in patients with CHF. Recently, spironolactone therapy has been reported to cause a decrease in the level of PIIINP in patients with CHF who have already received ACE inhibitors (25). Therefore, taken together with our findings, the significant reduction of mortality by spironolactone in CHF patients (8) who have already received furosemides and ACE inhibitors may be partly due to the inhibition of transcardiac ALDO extraction and to reduced collagen turnover in the failing heart. Study limitations. The transcardiac gradient of ALDO was indirect evidence of ALDO receptors in the heart, and the transcardiac extraction of circulating ALDO may not be simply due to the existence of intracellular mineralocorticoid receptors in the heart. Further studies are needed to evaluate the mechanism of ALDO extraction, but our data may be clinically important due to consistency with previous studies that the heart is one of the important target organs of ALDO. The small number of severe CHF patients (NYHA class IV) was also a limitation of the present study. Treatments, such as those using ACE inhibitors, beta-blockers and spironolactone, were not randomized. Despite these limitations of the effects of the present study, we can conclude that spironolactone therapy is an independent factor for reducing the transcardiac gradient of ALDO. We measured the plasma transcardiac gradient of ALDO before and after treatment with spironolactone in three patients with CHF, which was decreased after spironolactone in all three patients (data not shown). However, further studies including serial measurements of ALDO in the AO and CS and the left ventricular volume before and after treatment with spironolactone are needed to confirm our hypothesis.
Figure 3. Correlation between left ventricular end-diastolic volume index (LVEDVI) and the plasma levels of aldosterone (ALDO) in the aortic root (AO) and the transcardiac gradient of plasma ALDO in patients with congestive heart failure. AO ⫽ aortic root; CS ⫽ coronary sinus.
JACC Vol. 36, No. 3, 2000 September 2000:838–44
Tsutamoto et al. Aldosterone Extraction Through the Failing Heart
843
ALDO is extracted across the failing heart in CHF patients, and that therapy for reducing plasma ALDO and/or ALDO receptor antagonist would inhibit the process of left ventricular remodeling in patients with CHF. Acknowledgments We wish to thank Ms. Ikuko Sakaguchi for excellent technical assistance. We also express thanks to Mr. Daniel Mrozek for assistance in preparing the manuscript. Reprint requests and correspondence: Dr. Takayoshi Tsutamoto, First Department of Internal Medicine, Shiga University of Medical Science Tsukinowa, Seta, Otsu 520-2192, Japan. E-mail: [email protected].
REFERENCES
Figure 4. Correlation between the transcardiac gradient of plasma ALDO and the plasma levels of procollagen type III aminoterminal peptide (PIIINP) in coronary sinus (CS) in patients with congestive heart failure. AO ⫽ aortic root; CS ⫽ coronary sinus.
Conclusions. In both normal subjects and CHF patients, plasma ALDO was significantly lower in the CS than in the AO, suggesting ALDO extraction across the heart. The transcardiac gradient of plasma ALDO was independently regulated by plasma ALDO in the AO and by spironolactone therapy. Moreover, the transcardiac gradient of plasma ALDO was correlated with LVEDVI and plasma level of PIIINP, a marker of myocardial fibrosis in patients with CHF. These findings suggest that elevated circulating
Figure 5. Correlation between the plasma aldosterone (ALDO) level in the aortic root (AO) and the transcardiac gradient of plasma ALDO in 96 patients with congestive heart failure (CHF) who did not receive spironolactone. AO ⫽ aortic root; CS ⫽ coronary sinus. Open circles represent 96 CHF patients who did not receive spironolactone; closed circles represent 17 CHF patients who received spironolactone.
1. Cleland JGF, Dargie HJ, Hodsman GP, et al. Captopril in heart failure: a double-blind controlled trial. Br Heart J 1984;52:530 –5. 2. Staessen J, Lijnen P, Fagard R, Verschueren LJ, Amery A. Rise in plasma concentration of aldosterone during long-term angiotensin II supression. J Endocrinol 1965;25:219 –28. 3. Pitt D. ACE inhibitor co-therapy in patients with heart failure: rationale for the Randomized Aldactone Evaluation Study (RALES). Eur Heart J 1995;16 Suppl N:107–10. 4. Struthers AD. Aldosterone escape during ACE inhibitor therapy in chronic heart failure. Eur Heart J 1995;16 Suppl N:103–7. 5. Biollaz J, Brunner HR, Gavras I, Waeber B, Gavras H. Antihypertensive therapy with MK421: angiotensin II–renin relations to evaluate efficacy of converting enzyme blockade. J Cardiovasc Pharmacol 1982;4:966 –72. 6. Urata H, Healy B, Stewart R, Bumpus F, Husain A. Angiotensin II-forming pathways in normal and failing human hearts. Circ Res 1990;69:883–90. 7. Pitt B, Segal R, Martinez FA, et al. Randomised trial of losartan versus captopril in patients over 65 with heart failure. Lancet 1997;349:747– 52. 8. Pitt B, Zannad F, Remme WJ, et al. for the Randomized Aldactone Evaluation Study Investigators. The effect of spironolactone on mortality and morbidity in patients with severe heart failure. N Engl J Med 1999;341:709 –17. 9. Weber KT, Brilla C. Pathological hypertrophy and cardiac interstitium: fibrosis and renin–angiotensin–aldosterone system. Circulation 1991;83:1849 – 65. 10. Duprez DA, Bauwens FR, De Buyzere ML, et al. Influence of arterial blood pressure and aldosterone on left ventricular hypertrophy in moderate essential hypertension. Am J Cardiol 1993;71:17A–20A. 11. Navarro-Lopez F, Coca A, Pare JC, de la Sierra A, Bosch X, Urbano Marquez A. Left ventricular hypertrophy in asymptomatic essential hypertension: its relationship with aldosterone and the increase in sodium–proton exchanger activity. Eur Heart J 1993;14 Suppl J:38 – 41. 12. Schunkert H, Hense HW, Danser J, Muscholl M, Luncher A, Riegger AJG. Association between circulating components of the renin– angiotensin–aldosterone system and left ventricular mass. Br Heart J 1997;77:24 –31. 13. Scoot BA, Lawrence B, Nguyen HH, Meyer WJ III. Aldosterone and dexamethasone binding in human arterial smooth muscle cells. J Hypertens 1987;5:739 – 44. 14. Lombes M, Alfaidy N, Eugene E, Lessana A, Farman N, Bonvalet JP. Prerequisite for cardiac aldosterone action: mineralocorticoid receptor and 11-hydroxysteroid dehydrogenase in the human heart. Circulation 1995;92:175– 82. 15. Zennaro MC, Farman N, Bonbalet JP, Lombes M. Tissue-specific expression of ␣- and -messenger ribonucleic acid isoforms of the human mineralocorticoid receptor in normal and pathological states. J Clin Endocrinol Metab 1997;82:1345–52. 16. Young M, Fullerton M, Dilley R, Funder JW. Mineralocorticoids, hypertension, and cardiac fibrosis. J Clin Invest 1994;92:2578 – 83.
844
Tsutamoto et al. Aldosterone Extraction Through the Failing Heart
17. Brilla CG, Zhou G, Matsubara L, Weber KT. Collagen metabolism in cultured adult rat cardiac fibroblasts: response to angiotensin II and aldosterone. J Mol Cell Cardiol 1994;26:809 –20. 18. Lombes M, Oblin ME, Gasc JM, Baulieu EE, Farman N, Bonvalet JP. Immunohistochemical and biochemical evidence for a cardiovascular mineralocorticoid receptor. Circ Res 1992;71:503–10. 19. Robert V, Silvestre JS, Charlemagne D, et al. Biological determinants of aldosterone-induced cardiac fibrosis in rats. Hypertension 1995;26: 971– 8. 20. Silvestre JS, Robert V, Heymes C, et al. Myocardial production of aldosterone and corticosterone in the rat. J Biol Chem 1998;273:4883–91. 21. Tsutamoto T, Wada A, Maeda K, et al. Attenuation of compensation of endogenous cardiac natriuretic peptide system in chronic heart failure: prognostic role of plasma brain natriuretic peptide concentration in patients with chronic symptomatic left ventricular dysfunction. Circulation 1997;96:509 –16.
JACC Vol. 36, No. 3, 2000 September 2000:838–44 22. Young M, Funder JW. The renin–angiotensin–aldosterone system in experimental mineralocorticoid–salt–induced cardiac fibrosis. Am J Physiol 1996;271:E883– 8. 23. Klappacher G, Franzen P, Haab D, et al. Measuring extracellular matrix turnover in the serum of patients with idiopathic or ischemic dilated cardiomyopathy and impact on diagnosis and prognosis. Am J Cardiol 1995;75:913– 8. 24. Host NB, Jensen LT, Bendixen PM, Jensen SE, Koldkjaer OG, Simonsen EE. The aminoterminal propeptide of type III procollagen provides new information on prognosis after acute myocardial infarction. Am J Cardiol 1995;76:869 –73. 25. MacFadyen RJ, Barr CS, Struthers AD. Aldosterone blockade reduces vascular collagen turnover, improves heart rate variability and reduces early morning rise in heart rate in heart failure patients. Cardiovasc Res 1997;35:30 – 4.
Vol. 36, No. 3, 2000 ISSN 0735-1097/00/$20.00 PII S0735-1097(00)00796-8
Spironolactone Inhibits the Transcardiac Extraction of Aldosterone in Patients With Congestive Heart Failure Takayoshi Tsutamoto, MD, Atsuyuki Wada, MD, Keiko Maeda, MD, Naoko Mabuchi, MD, Masaru Hayashi, MD, Takashi Tsutsui, MD, Masato Ohnishi, MD, Masahide Sawaki, MD, Masanori Fujii, MD, Takehiro Matsumoto, MD, Hajime Horie, MD, Yoshihisa Sugimoto, MD, Masahiko Kinoshita, MD Otsu, Japan The study evaluated the transcardiac extraction or spillover of aldosterone (ALDO) in normal subjects and in patients with congestive heart failure (CHF). BACKGROUND Aldosterone promotes collagen synthesis and structural remodeling of target organs such as the heart. Spironolactone, an ALDO receptor antagonist, has recently been reported to reduce the mortality of patients with CHF; however, the effects of spironolactone on the transcardiac gradient of ALDO have not been clarified. METHODS We measured plasma ALDO in the aortic root (AO) and coronary sinus (CS) in normal subjects and 113 consecutive CHF patients and also measured plasma procollagen type III aminoterminal peptide (PIIINP) in CS, a biochemical marker of myocardial fibrosis. RESULTS Plasma ALDO was significantly lower in the CS than in the AO in normal subjects (n ⫽ 15; 61.2 ⫾ 9.3 vs. 83.1 ⫾ 11.8 pg/ml, p ⬍ 0.0001). In 96 CHF patients who did not receive spironolactone, plasma ALDO was significantly lower in the CS than in the AO (59.3 ⫾ 3.9 vs. 73.8 ⫾ 4.9 pg/ml, p ⬍ 0.0001). In contrast to the difference in these 96 patients, there was no significant difference in ALDO between the AO and CS in 17 patients who received spironolactone (127.4 ⫾ 20 vs. 124.0 ⫾ 19 pg/ml, p ⫽ 0.50). Stepwise multivariate analyses showed that spironolactone therapy had an independent and significant negative relationship with the transcardiac gradient of plasma ALDO in patients with CHF. In addition, significant positive correlations were seen between the transcardiac gradient of plasma ALDO and PIIINP (r ⫽ 0.565, p ⬍ 0.0001) and the left ventricular end-diastolic volume index (r ⫽ 0.484, p ⬍ 0.0001). CONCLUSIONS These results indicate that plasma ALDO is extracted through the heart in normal subjects and in CHF patients who do not receive spironolactone and that spironolactone inhibits the transcardiac extraction of ALDO in CHF patients, suggesting that spironolactone blocks the effects of ALDO on the failing heart in patients with CHF. (J Am Coll Cardiol 2000;36: 838 – 44) © 2000 by the American College of Cardiology OBJECTIVES
Despite the outstanding success of angiotensin-converting enzyme (ACE) inhibitors in the treatment of congestive heart failure (CHF), the mortality rate of CHF patients remains higher than that of normal subjects. This high mortality rate is partly due to the aldosterone escape phenomenon in patients who received ACE inhibitors (1– 4). Moreover, the plasma angiotensin II (Ang II) level is not constantly decreased during long-term treatment with ACE inhibitors (5), partly because Ang II may be produced by alternative pathways (6). Therefore, angiotensin type 1 receptor antagonists or aldosterone (ALDO) receptor antagonists such as spironolactone may have additional advantages over ACE inhibitors (3,7). Recently, spironolactone was shown to reduce the mortality of CHF patients who had already received ACE inhibitors, digitalis, and diuretics. In the Randomized Aldactone Evaluation Study (RALES) (3,8), both sudden death and CHF death were significantly reduced with spironolactone. However, despite the great impact of the RALES results, the mechanism of the From the First Department of Internal Medicine, Shiga University of Medical Science, Tsukinowa, Seta, Otsu, Japan. Supported in part by Grants-in-Aid for Scientific Research of Japan. Manuscript received December 8, 1999; revised manuscript received March 15, 2000, accepted April 26, 2000.
improvement of mortality with spironolactone remains to be fully elucidated. Aldosterone displays both myocardial and renal effects that may have profound implications for left ventricular remodeling (9). Significant associations were found between plasma ALDO and left ventricular mass in patients with hypertension as well as in a population-based sample, suggesting that aldosterone may be an important factor in the modulation of cardiac structure (10 –12). Recently, it has been reported that mineralocorticoid receptor, which mediates the action of ALDO, is expressed in cardiomyocytes, endothelial cells, and fibroblasts in the human heart (13–15). Aldosterone, secreted from the adrenal gland, has been shown to stimulate cardiac collagen synthesis and fibroblast proliferation via activation of local mineralocorticoid receptors or via an unexplained indirect mechanism (16 –19). However, the relationship between the cardiac effect of ALDO and the mechanism of cardiac fibrosis by ALDO has not been fully elucidated. More recently, cardiac production of ALDO has been reported in rats (20), suggesting the possibility of cardiac production of ALDO in humans. However, a transcardiac gradient of ALDO in humans has not been demonstrated. In the present study, we measured plasma levels of ALDO both in the aortic root and the coronary sinus, and
Tsutamoto et al. Aldosterone Extraction Through the Failing Heart
JACC Vol. 36, No. 3, 2000 September 2000:838–44
Abbreviations and Acronyms ACE ⫽ angiotensin-converting enzyme ALDO ⫽ aldosterone Ang II ⫽ angiotensin II ANP ⫽ atrial natriuretic peptide AO ⫽ aortic root BNP ⫽ brain natriuretic peptide CHF ⫽ congestive heart failure CS ⫽ coronary sinus LVEDVI ⫽ left ventricular end-diastolic volume index LVEF ⫽ left ventricular ejection fraction PIIINP ⫽ procollagen type III aminoterminal peptide
we evaluated whether ALDO is extracted or spilled over through the heart in normal subjects and in patients with CHF. Moreover, we also evaluated the effects of spironolactone on the transcardiac gradient of ALDO and determined the relationship between the transcardiac gradient of ALDO and the left ventricular remodeling.
METHODS Patients. The subjects were 113 consecutive patients with symptomatic left ventricular dysfunction (left ventricular ejection fraction [LVEF] ⬍45%) without renal failure who underwent cardiac catheterization for clinical indications. Patients with angina pectoris, secondary hypertensive heart disease, renal failure, or liver dysfunction were excluded. Patients with coronary stenosis (⬎70%) were excluded. We also selected 15 age-matched normal subjects (age, 34 to 71; mean: 54 years) who were admitted complaining of chest pain and whose hearts proved to be normal by coronary angiography. Informed consent was obtained from all patients before participation in the study, and the protocol was approved by the Human Investigations Committee of our institution. The subjects were 75 men and 38 women ranging in age from 17 to 79 years (mean: 58 years); 68 patients had suffered a myocardial infarction more than three months before the study, 38 had dilated cardiomyopathy, 5 had hypertensive heart disease, and 2 had valvular heart disease. Sixty-eight patients with old myocardial infarction had previously received percutaneous transluminal coronary angioplasty (PTCA) in our hospital. Eighty-four patients were classified according to the standards of the New York Heart Association (NYHA) as functional class II, 25 patients as class III, and 4 patients as
839
class IV. At entry to the study, 76 patients were treated with furosemide, 17 with spironolactone (a dose of 25 to 50 mg daily), 79 with ACE inhibitors, 58 with digitalis, 37 with vasodilators, and 27 with beta-blockers. Most drugs had been administered for more than two months. Study protocol. All patients were premedicated with an oral dose of diazepam (5 mg) and rested in bed in the supine position for at least 20 min. Right-sided cardiac catheterization was performed using a 7F Swan-Ganz catheter. The heart rate was monitored by electrocardiography. Blood samples for measuring plasma ALDO were collected simultaneously from the aortic root (AO) and coronary sinus (CS). A 6F catheter (Goodman, Tokyo, Japan) for blood sampling was positioned in the CS, and the position of the catheter was confirmed by injection of contrast medium just after blood sampling. The validity of CS blood sampling was also verified by simultaneous measurements of the atrial natriuretic peptide (ANP) and brain natriuretic peptide (BNP) levels in the AO and CS. In 79 of 113 patients, blood samples for measuring the plasma levels of procollagen type III aminoterminal peptide (PIIINP) were drawn from the CS. A Swan-Ganz catheter was inserted through the right femoral vein into the main pulmonary artery, where the pressure was measured. The catheter was then advanced into the pulmonary artery, and the pulmonary capillary wedge pressure was measured by inflating the balloon. Cardiac output was determined by the thermodilution method immediately after blood collection. Left ventriculography was performed using contrast medium after the hemodynamic measurements and blood sampling. Measurement of neurohumoral factors. Blood for the measurement of the plasma levels of ANP and BNP was transferred to a chilled tube containing EDTA (1 mg/ml) and aprotinin (500 kallikrein inactivator units/ml), and then centrifuged at 3,000 rpm for 15 min at 4°C. The plasma thus obtained was stored at ⫺30°C until assayed. Plasma concentrations of ANP and BNP were measured with a specific immunoradiometric assay for using a commercial kit (Shionogi, Osaka, Japan) as previously reported (21). Blood for measurement of the plasma levels of ALDO was transferred to a chilled tube containing EDTA (1 mg/ml), centrifuged at 3,000 rpm for 15 min at 4°C, and the plasma thus obtained was stored at ⫺30°C until it was assayed. Plasma ALDO levels were measured using a commercial radioimmunoassay kit. This assay system for
Table 1. Hemodynamic and Neurohumoral Data
Mild CHF (n ⫽ 84) (SE) Severe CHF (n ⫽ 29) (SE)
LVEF (%)
MPAP (mm Hg)
PCWP (mm Hg)
LVEDP (mm Hg)
LVEDVI (l/min/m2)
ANP (pg/ml)
BNP (pg/ml)
ALDO in AO (pg/ml)
ALDO in CS (pg/ml)
35.6 (0.7) 22.9* (1.8)
16.9 (0.4) 30.6* (2.1)
8.7 (0.3) 20.7* (1.8)
11.3 (0.5) 25.4* (1.9)
122 (3.9) 196* (12)
71.7 (11.3) 207* (24)
87.5 (8.3) 459* (72)
73.6 (5.6) 119.6* (15)
61.2 (4.4) 102* (14)
LVEF ⫽ left ventricular ejection fraction; MPAP ⫽ mean pulmonary arterial pressure; PCWP ⫽ pulmonary capillary wedge pressure; LVEDP ⫽ left ventricular end-diastolic pressure; LVEDVI ⫽ left ventricular end-diastolic volume index; ANP ⫽ atrial natriuretic peptide; BNP ⫽ brain natriuretic peptide; ALDO ⫽ aldosterone; AO ⫽ ascending aorta; CS ⫽ coronary sinus; SE ⫽ standard error. *p ⬍ 0.0001 vs. mild CHF.
840
Tsutamoto et al. Aldosterone Extraction Through the Failing Heart
JACC Vol. 36, No. 3, 2000 September 2000:838–44
Table 2. Univariate and Multivariate Linear Model of Left Ventricular End-Diastolic Volume Index in 113 Patients With CHF
Variable Age (yr) Gender (male ⫽ 1; female ⫽ 0) Etiology (ICM ⫽ 1; non-ICM ⫽ 0) NYHA functional class LVEF (%) Mean blood pressure (mm Hg) Cardiac index (l/min/m2) Left ventricular end-diastolic pressure (mm Hg) Atrial natriuretic peptide (pg/ml) Brain natriuretic peptide (pg/ml) ALDO in AO (pg/ml) (AO-CS) ALDO (pg/ml) Treatment (treatment ⫽ 1; no treatment ⫽ 0) Spironolactone Diuretics Digitalis ACE inhibitors Vasodilators Beta-blockers
Univariate Correlation Coefficient
p Value
⫺0.115 ⫺0.223 ⫺0.536 0.473 ⫺0.665 ⫺0.236 ⫺0.029 0.337 0.487 0.351 0.317 0.484
0.283 0.0057 ⬍0.0001 ⬍0.0001 ⬍0.0001 0.003 0.758 0.0003 ⬍0.0001 0.0002 0.0006 ⬍0.0001
0.302 0.473 0.419 ⫺0.174 ⫺0.290 0.117
0.005 ⬍0.0001 ⬍0.0001 0.081 0.006 0.128
Multivariate Beta Coefficient
⫺3.4
0.720
p Value
⬍0.0001
⬍0.0001
ICM ⫽ ischemic cardiomyopathy; LVEF ⫽ left ventricular ejection fraction; ALDO ⫽ plasma aldosterone; AO ⫽ aortic root; CS ⫽ coronary sinus; ACE ⫽ angiotensin-converting enzyme.
plasma ALDO did not cross-react with angiotensin I or II, spironolactone, ANP, or BNP. Plasma levels of PIIINP were measured with a specific immunoradiometric assay using a commercial kit (CIS Bio International, Nagoya, Japan).
Statistical analysis. All results are expressed as the mean ⫾ SEM. Univariate analyses were performed using the Student t test. To evaluate the contribution of ALDO to left ventricular remodeling, univariate and stepwise multivariate analyses were used among the 18 variables. Univariate and
Table 3. Univariate and Multivariate Linear Model of Transcardiac Gradient of Plasma Aldosterone in 113 Patients With CHF
Variable Age (yrs) Gender (male ⫽ 1; female ⫽ 0) Etiology (ICM ⫽ 1, non-ICM ⫽ 0) NYHA functional class LVEF (%) Mean blood pressure (mm Hg) Cardiac index (l/min/m2) Left ventricular end-diastolic blood pressure (mm Hg) Left ventricular end-diastolic volume index (ml/m2) Atrial natriuretic peptide (pg/ml) Brain natriuretic peptide (pg/ml) ALDO (pg/ml) Treatment (treatment ⫽ 1; no treatment ⫽ 0) Spironolactone Diuretics Digitalis ACE inhibitors Vasodilators Beta-blockers
Univariate Correlation Coefficient
p Value
⫺0.031 ⫺0.067 ⫺0.109 0.280 ⫺0.224 ⫺0.180 ⫺0.097 0.250
0.744 0.438 0.251 0.0029 0.018 0.059 0.309 0.008
0.484
⬍0.0001
0.217
0.370 0.231 0.371
⬍0.0001 0.389 ⬍0.0001
0.10
⫺0.207 0.114 0.056 ⫺0.176 ⫺0.089 ⫺0.062
0.027 0.228 0.554 0.062 0.348 0.512
Multivariate Beta Coefficient
⫺14.7
p Value
⬍0.0001
0.0052
0.0115
ICM ⫽ ischemic cardiomyopathy; LVEF ⫽ left ventricular ejection fraction; ALDO ⫽ plasma aldosterone; AO ⫽ aortic root; ACE ⫽ angiotensin-converting enzyme.
JACC Vol. 36, No. 3, 2000 September 2000:838–44
Figure 1. Plasma aldosterone (ALDO) concentrations in the aortic root (AO) and coronary sinus (CS) in normal subjects.
stepwise multivariate linear regression analyses were also used to detect independent predictors of the transcardiac gradient of plasma ALDO among the 18 variables. Linear regression analysis was used to determine the relationship between continuous variables. A p value ⬍0.05 was regarded as significant.
RESULTS Difference between plasma concentration of ALDO in aortic root and coronary sinus. In the 15 age-matched normal subjects, the plasma ALDO level was significantly lower in the CS than in the AO (Fig. 1) (61.2 ⫾ 9.3 vs. 83.1 ⫾ 11.8 pg/ml, p ⬍ 0.0001). In 113 consecutive CHF patients, plasma ALDO was also significantly lower in the CS than in the AO (69.0 ⫾ 4.8 vs. 81.9 ⫾ 5.4 pg/ml, p ⬍ 0.0001). In the 96 patients without spironolactone treatment, plasma ALDO was significantly lower in the CS than in the AO (Fig. 2) (59.3 ⫾ 3.9 vs. 73.8 ⫾ 4.9 pg/ml, p ⬍ 0.0001). In contrast to the difference in the patients who did not receive spironolactone, there was no significant difference of ALDO between the AO and CS in the 17 patients who received spironolactone (Fig. 2) (127.4 ⫾ 20 vs. 124.0 ⫾ 19 pg/ml, p ⫽ 0.50). Plasma ALDO level was significantly higher in the patients who received spironolactone than in the patients who did not receive spironolactone. Hemodynamic and neurohumoral data. Patients were divided into two groups according to symptoms and were analyzed for hemodynamic and neurohumoral data (Table 1). Hemodynamic parameters such as pulmonary capillary wedge pressure, left ventricular end-diastolic pressure, and left ventricular end-diastolic volume index (LVEDVI) were significantly higher in severe CHF patients (NYHA class III or class IV) than in mild CHF patients (NYHA class II). The LVEF was significantly lower in severe CHF than in mild CHF patients. Neurohumoral factors such as ANP, BNP, and ALDO were significantly higher in severe CHF patients than in mild CHF patients. Plasma ALDO was
Tsutamoto et al. Aldosterone Extraction Through the Failing Heart
841
Figure 2. Plasma aldosterone (ALDO) concentrations in the aortic root (AO) and coronary sinus (CS) in patients with congestive heart failure (CHF). Left panel shows data of ALDO in 96 CHF patients who did not receive spironolactone; right panel shows data of ALDO in 17 CHF patients who received spironolactone. *, p ⬍ 0.0001 vs. the value of AO.
significantly lower in the CS than in the AO in mild CHF patients and in severe CHF patients. Relationship between plasma ALDO extraction across the heart and LVEDVI. Table 2 shows the results of univariate and multivariate analyses to assess the factors regulating the LVEDVI in 113 patients with CHF. According to stepwise multivariate analyses, only a high transcardiac gradient of plasma ALDO (p ⬍ 0.0001) and LVEF (p ⬍ 0.0001) were significant independent predictors of a large LVEDVI. Figure 3 shows the correlation between the LVEDVI and the plasma levels of ALDO in the AO (r ⫽ 0.317; p ⫽ 0.0006) and the transcardiac gradient of plasma ALDO (r ⫽ 0.484; p ⬍ 0.0001) in patients with CHF. A significant positive correlation existed between the transcardiac extraction of plasma ALDO and plasma level of PIIINP in the CS (Fig. 4; r ⫽ 0.565; p ⬍ 0.0001), and also a significant correlation between the plasma level of PIIINP in the CS and LVEDVI (r ⫽ 0.448; p ⬍ 0.0001). Relationship between plasma ALDO in aortic root and ALDO extraction across the heart. Table 3 shows the results of univariate and multivariate analyses among 18 variables to assess the factors regulating the transcardiac extraction of plasma ALDO. According to stepwise multivariate analyses, a high level of plasma ALDO in the AO and treatment with spironolactone were significant independent predictors of transcardiac extraction of plasma ALDO in patients with CHF. Figure 5 shows a significant correlation between the plasma level of ALDO in the AO and the transcardiac extraction of plasma ALDO by linear regression analysis in the 96 patients not treated with spironolactone (r ⫽ 0.759; p ⬍ 0.0001). There was no correlation between the plasma level of ALDO in the AO and the transcardiac extraction of plasma ALDO in 17 patients who received spironolactone, and most of the 17 patients who received spironolactone were on the right side of the linear regression line of the 96 patients who did not receive spironolactone (Fig. 5).
842
Tsutamoto et al. Aldosterone Extraction Through the Failing Heart
DISCUSSION It has been recently reported that mineralocorticoid receptors, which mediate the action of ALDO, are expressed in the human heart (13–15). Cardiac fibrosis in rats induced by ALDO infusion was not inhibited by concomitant administration of an ACE inhibitor or an Ang II receptor antagonist, losartan, but was completely blocked by concomitant administration of the mineralocorticoid receptor antagonist potassium canrenoate (22). However, the relationship between the cardiac effect of ALDO and the mechanism of cardiac fibrosis by ALDO has not been fully elucidated in vivo. We demonstrated here for the first time that plasma ALDO is extracted through the heart in normal subjects and in patients with CHF, and that a positive correlation exists between the transcardiac gradient of ALDO and LVEDVI. Moreover, spironolactone therapy inhibited the ALDO extraction through the failing heart. These findings indicate that the heart is a target organ of ALDO, and that spironolactone, a mineralocorticoid receptor antagonist, inhibits the ALDO extraction through the heart in patients with CHF. We also found a positive correlation between the transcardiac ALDO extraction and the plasma level of PIIINP, a possible biochemical marker of myocardial fibrosis (23,24) in the CS, suggesting that the increase of the transcardiac ALDO extraction in the failing heart stimulates myocardial collagen turnover, as shown in vitro (19). Our finding of a positive correlation between the transcardiac ALDO extraction and the plasma level of PIIINP also suggests that the sustained transcardiac ALDO extraction is an important modulator of left ventricular remodeling in the failing heart regardless of the plasma ALDO levels. Klappacher et al. (23) reported a significant positive correlation between serum PIIINP and the amount of myocardial collagen type III on cardiac biopsy in patients with CHF. Moreover, Host et al. (24) showed that after a myocardial infarction, the plasma level of PIIINP was higher in those patients with a poor prognosis. Therefore, the plasma level of PIIINP may be a biochemical marker of myocardial fibrosis and/or left
JACC Vol. 36, No. 3, 2000 September 2000:838–44
ventricular remodeling in patients with CHF. Indeed, in the present study, there was a positive correlation between the plasma level of PIIINP and LVEDVI. In addition, the positive correlation between the transcardiac ALDO extraction and the plasma level of PIIINP suggests an interaction between ALDO extraction and left ventricular remodeling in patients with CHF. Recently, spironolactone therapy has been reported to cause a decrease in the level of PIIINP in patients with CHF who have already received ACE inhibitors (25). Therefore, taken together with our findings, the significant reduction of mortality by spironolactone in CHF patients (8) who have already received furosemides and ACE inhibitors may be partly due to the inhibition of transcardiac ALDO extraction and to reduced collagen turnover in the failing heart. Study limitations. The transcardiac gradient of ALDO was indirect evidence of ALDO receptors in the heart, and the transcardiac extraction of circulating ALDO may not be simply due to the existence of intracellular mineralocorticoid receptors in the heart. Further studies are needed to evaluate the mechanism of ALDO extraction, but our data may be clinically important due to consistency with previous studies that the heart is one of the important target organs of ALDO. The small number of severe CHF patients (NYHA class IV) was also a limitation of the present study. Treatments, such as those using ACE inhibitors, beta-blockers and spironolactone, were not randomized. Despite these limitations of the effects of the present study, we can conclude that spironolactone therapy is an independent factor for reducing the transcardiac gradient of ALDO. We measured the plasma transcardiac gradient of ALDO before and after treatment with spironolactone in three patients with CHF, which was decreased after spironolactone in all three patients (data not shown). However, further studies including serial measurements of ALDO in the AO and CS and the left ventricular volume before and after treatment with spironolactone are needed to confirm our hypothesis.
Figure 3. Correlation between left ventricular end-diastolic volume index (LVEDVI) and the plasma levels of aldosterone (ALDO) in the aortic root (AO) and the transcardiac gradient of plasma ALDO in patients with congestive heart failure. AO ⫽ aortic root; CS ⫽ coronary sinus.
JACC Vol. 36, No. 3, 2000 September 2000:838–44
Tsutamoto et al. Aldosterone Extraction Through the Failing Heart
843
ALDO is extracted across the failing heart in CHF patients, and that therapy for reducing plasma ALDO and/or ALDO receptor antagonist would inhibit the process of left ventricular remodeling in patients with CHF. Acknowledgments We wish to thank Ms. Ikuko Sakaguchi for excellent technical assistance. We also express thanks to Mr. Daniel Mrozek for assistance in preparing the manuscript. Reprint requests and correspondence: Dr. Takayoshi Tsutamoto, First Department of Internal Medicine, Shiga University of Medical Science Tsukinowa, Seta, Otsu 520-2192, Japan. E-mail: [email protected].
REFERENCES
Figure 4. Correlation between the transcardiac gradient of plasma ALDO and the plasma levels of procollagen type III aminoterminal peptide (PIIINP) in coronary sinus (CS) in patients with congestive heart failure. AO ⫽ aortic root; CS ⫽ coronary sinus.
Conclusions. In both normal subjects and CHF patients, plasma ALDO was significantly lower in the CS than in the AO, suggesting ALDO extraction across the heart. The transcardiac gradient of plasma ALDO was independently regulated by plasma ALDO in the AO and by spironolactone therapy. Moreover, the transcardiac gradient of plasma ALDO was correlated with LVEDVI and plasma level of PIIINP, a marker of myocardial fibrosis in patients with CHF. These findings suggest that elevated circulating
Figure 5. Correlation between the plasma aldosterone (ALDO) level in the aortic root (AO) and the transcardiac gradient of plasma ALDO in 96 patients with congestive heart failure (CHF) who did not receive spironolactone. AO ⫽ aortic root; CS ⫽ coronary sinus. Open circles represent 96 CHF patients who did not receive spironolactone; closed circles represent 17 CHF patients who received spironolactone.
1. Cleland JGF, Dargie HJ, Hodsman GP, et al. Captopril in heart failure: a double-blind controlled trial. Br Heart J 1984;52:530 –5. 2. Staessen J, Lijnen P, Fagard R, Verschueren LJ, Amery A. Rise in plasma concentration of aldosterone during long-term angiotensin II supression. J Endocrinol 1965;25:219 –28. 3. Pitt D. ACE inhibitor co-therapy in patients with heart failure: rationale for the Randomized Aldactone Evaluation Study (RALES). Eur Heart J 1995;16 Suppl N:107–10. 4. Struthers AD. Aldosterone escape during ACE inhibitor therapy in chronic heart failure. Eur Heart J 1995;16 Suppl N:103–7. 5. Biollaz J, Brunner HR, Gavras I, Waeber B, Gavras H. Antihypertensive therapy with MK421: angiotensin II–renin relations to evaluate efficacy of converting enzyme blockade. J Cardiovasc Pharmacol 1982;4:966 –72. 6. Urata H, Healy B, Stewart R, Bumpus F, Husain A. Angiotensin II-forming pathways in normal and failing human hearts. Circ Res 1990;69:883–90. 7. Pitt B, Segal R, Martinez FA, et al. Randomised trial of losartan versus captopril in patients over 65 with heart failure. Lancet 1997;349:747– 52. 8. Pitt B, Zannad F, Remme WJ, et al. for the Randomized Aldactone Evaluation Study Investigators. The effect of spironolactone on mortality and morbidity in patients with severe heart failure. N Engl J Med 1999;341:709 –17. 9. Weber KT, Brilla C. Pathological hypertrophy and cardiac interstitium: fibrosis and renin–angiotensin–aldosterone system. Circulation 1991;83:1849 – 65. 10. Duprez DA, Bauwens FR, De Buyzere ML, et al. Influence of arterial blood pressure and aldosterone on left ventricular hypertrophy in moderate essential hypertension. Am J Cardiol 1993;71:17A–20A. 11. Navarro-Lopez F, Coca A, Pare JC, de la Sierra A, Bosch X, Urbano Marquez A. Left ventricular hypertrophy in asymptomatic essential hypertension: its relationship with aldosterone and the increase in sodium–proton exchanger activity. Eur Heart J 1993;14 Suppl J:38 – 41. 12. Schunkert H, Hense HW, Danser J, Muscholl M, Luncher A, Riegger AJG. Association between circulating components of the renin– angiotensin–aldosterone system and left ventricular mass. Br Heart J 1997;77:24 –31. 13. Scoot BA, Lawrence B, Nguyen HH, Meyer WJ III. Aldosterone and dexamethasone binding in human arterial smooth muscle cells. J Hypertens 1987;5:739 – 44. 14. Lombes M, Alfaidy N, Eugene E, Lessana A, Farman N, Bonvalet JP. Prerequisite for cardiac aldosterone action: mineralocorticoid receptor and 11-hydroxysteroid dehydrogenase in the human heart. Circulation 1995;92:175– 82. 15. Zennaro MC, Farman N, Bonbalet JP, Lombes M. Tissue-specific expression of ␣- and -messenger ribonucleic acid isoforms of the human mineralocorticoid receptor in normal and pathological states. J Clin Endocrinol Metab 1997;82:1345–52. 16. Young M, Fullerton M, Dilley R, Funder JW. Mineralocorticoids, hypertension, and cardiac fibrosis. J Clin Invest 1994;92:2578 – 83.
844
Tsutamoto et al. Aldosterone Extraction Through the Failing Heart
17. Brilla CG, Zhou G, Matsubara L, Weber KT. Collagen metabolism in cultured adult rat cardiac fibroblasts: response to angiotensin II and aldosterone. J Mol Cell Cardiol 1994;26:809 –20. 18. Lombes M, Oblin ME, Gasc JM, Baulieu EE, Farman N, Bonvalet JP. Immunohistochemical and biochemical evidence for a cardiovascular mineralocorticoid receptor. Circ Res 1992;71:503–10. 19. Robert V, Silvestre JS, Charlemagne D, et al. Biological determinants of aldosterone-induced cardiac fibrosis in rats. Hypertension 1995;26: 971– 8. 20. Silvestre JS, Robert V, Heymes C, et al. Myocardial production of aldosterone and corticosterone in the rat. J Biol Chem 1998;273:4883–91. 21. Tsutamoto T, Wada A, Maeda K, et al. Attenuation of compensation of endogenous cardiac natriuretic peptide system in chronic heart failure: prognostic role of plasma brain natriuretic peptide concentration in patients with chronic symptomatic left ventricular dysfunction. Circulation 1997;96:509 –16.
JACC Vol. 36, No. 3, 2000 September 2000:838–44 22. Young M, Funder JW. The renin–angiotensin–aldosterone system in experimental mineralocorticoid–salt–induced cardiac fibrosis. Am J Physiol 1996;271:E883– 8. 23. Klappacher G, Franzen P, Haab D, et al. Measuring extracellular matrix turnover in the serum of patients with idiopathic or ischemic dilated cardiomyopathy and impact on diagnosis and prognosis. Am J Cardiol 1995;75:913– 8. 24. Host NB, Jensen LT, Bendixen PM, Jensen SE, Koldkjaer OG, Simonsen EE. The aminoterminal propeptide of type III procollagen provides new information on prognosis after acute myocardial infarction. Am J Cardiol 1995;76:869 –73. 25. MacFadyen RJ, Barr CS, Struthers AD. Aldosterone blockade reduces vascular collagen turnover, improves heart rate variability and reduces early morning rise in heart rate in heart failure patients. Cardiovasc Res 1997;35:30 – 4.