Author + information
- Received January 24, 2007
- Revision received March 9, 2007
- Accepted March 12, 2007
- Published online August 14, 2007.
- Anna K.Y. Chan, MB, MRCP⁎,
- John E. Sanderson, MD, FRCP, FACC⁎,⁎ (, )
- Tian Wang, PhD⁎,
- Wynnie Lam, FRCR†,
- Gabriel Yip, MD⁎,
- Mei Wang, MD, PhD⁎,
- Yat-Yin Lam, MB, MRCP⁎,
- Yan Zhang, PhD⁎,
- Leata Yeung, RN, MPhil⁎,
- Eugene B. Wu, MD⁎,
- Wilson W.M. Chan, MD⁎,
- John T.H. Wong, MB, MRCP⁎,
- Nina So, FRCR† and
- Cheuk-Man Yu, MD, FRCP⁎
- ↵⁎Reprint requests and correspondence:
Prof. John E. Sanderson, Department of Cardiovascular Medicine, The Medical School, University of Birmingham, Edgbaston, Birmingham B15 2TT, United Kingdom.
Objectives The objective of this study was to determine if adding spironolactone to an angiotensin II receptor blocker improves left ventricular (LV) function, mass, and volumes in chronic heart failure.
Background Add-on spironolactone therapy substantially improves clinical outcomes among patients with severe heart failure (HF) on standard therapy. However, the value of combining spironolactone with an angiotensin II receptor blocker on LV reverse remodeling in mild-to-moderate systolic HF is unclear.
Methods Fifty-one systolic HF patients with left ventricular ejection fraction (LVEF) <40% were randomly assigned to receive 1-year treatment of candesartan and spironolactone (combination group) or candesartan and placebo (control group). Reverse remodeling was assessed by serial cardiac magnetic resonance imaging and echocardiographic tissue Doppler imaging (TDI).
Results There were significant improvements in LVEF (35 ± 3% vs. 26 ± 2%, p < 0.01) and reduction of LV end-diastolic volume index (121 ± 16 ml/m2vs. 155 ± 14 ml/m2, p = 0.001), end-systolic volume index (88 ± 17 ml/m2vs. 120 ± 15 ml/m2, p < 0.0005), and LV mass index (81 ± 6 g/m2vs. 93 ± 6 g/m2, p = 0.002) in the combination group at 1 year. In addition, there was significant increase in peak basal systolic velocity and strain by TDI, decrease in index of filling pressure, and increase in cyclic variation integrated backscatter. In the control group, there were no significant changes in all these parameters after 1 year.
Conclusions The addition of spironolactone to candesartan has significant beneficial effects on LV reverse remodeling in patients with mild-to-moderate chronic systolic HF.
Excess aldosterone is well known to have diverse deleterious effects in systolic heart failure (HF) patients (1). Previous studies have established the beneficial role of the aldosterone receptor blocker spironolactone in treating patients with severe HF (2), and there is evidence suggesting that aldosterone “breakthrough” occurs that might attenuate the cardioprotective effects of long-term angiotensin-converting enzyme (ACE) inhibitor monotherapy (3,4). Similar concerns regarding aldosterone “breakthrough” with angiotensin II receptor blocker (ARB) monotherapy have been raised, although the results from previous studies are conflicting (5,6). In fact, whether spironolactone might exert an additive beneficial effect to that of ARB on left ventricular (LV) function and reverse remodeling in patients with mild-to-moderate systolic HF is unknown.
Therefore, we investigated the additive beneficial effects of combining candesartan with spironolactone on reverse LV remodeling with serial cardiac magnetic resonance imaging (CMR) and tissue Doppler imaging (TDI).
Study protocol and randomization
This was a prospective, randomized, double-blind, placebo-controlled study. Eligible patients with left ventricular ejection fraction (LVEF) <40% and receiving ACE inhibitors more than 6 months were randomly assigned to replace ACE inhibitors with candesartan 8 mg and spironolactone 25 mg once daily (combination group) or candesartan 8 mg and a matching identical placebo once daily (control group) over a 1-year study period. According to the study protocol, enrolled patients underwent serial CMR, echocardiography, and clinical assessments and laboratory tests (Fig. 1).
Eligible patients were receiving standard anti-HF treatment except for potassium-sparing diuretics before randomization. Dosages of these background medications except diuretics were not allowed to change after randomization. Patients were excluded if they had significant valvular heart disease; congenital heart disease; any life-threatening disease with limited life expectancy; or standard contraindications for CMR examination, creatinine concentration >200 μmol/l, potassium level >5 mmol/l, and a history of allergy or side-effect with spironolactone. All patients gave written informed consent. The institutional ethics committee approved the study protocol.
All CMR were performed with a 1.5-T scanner (Sonata, Siemens, Erlangen, Germany). Cine imaging with steady state precession sequence were performed. Raw images were processed by manual outlining of the endocardial and epicardial borders at end-diastolic and end-systolic frames. Analysis of LVEF, left ventricular end-systolic volume index (LVESVI) and left ventricular end-diastolic volume index (LVEDVI), and LV mass indexes by CMR were calculated with Simpson’s rule.
TDI and strain imaging by echocardiography
Echocardiograms were performed with a standard ultrasound machine (Vivid 5 and Vivid 7, GE Vingmed, Horten, Norway). The TDI and strain imaging were acquired as previously described (7). Myocardial longitudinal velocity and deformation curves were obtained, and peak systolic velocity (Sm) or strain during the ejection phase as well as early diastolic velocity (Em) were measured. The filling pressure was estimated by the ratio of transmitral early diastolic Doppler velocity (E) to Em measured at the basal septal segment (E/Em) (8).
Integrated backscatter imaging
Integrated backscatter imaging was acquired with parasternal long-axis view as previously described (9). The magnitude of cyclic variation of integrated backscatter (CVIB) was defined as the difference between the minimal and maximal values in a cardiac cycle. The final data were presented as mean value of the septum and posterior wall.
The study had a 90% power to detect an 8% difference in LVEF and volume with a minimum of 20 subjects in each group with 2-sided alpha of 0.05, assuming an SD of 8% for the CMR measurement of LVEF and LV volumes. All continuous values were expressed as mean ± SE. Comparisons of the baseline characteristics between groups were performed with the Pearson chi-square test for categorical variables. Comparisons within treatment group for CMR volumetric variables at baseline and 26 and 52 weeks were performed by paired ttest. A repeated measure analysis of variance (ANOVA) for trends were used to assess treatment effect on variables between treatment groups related to volumetric changes and ventricular remodeling. A value of p < 0.05 was considered statistically significant. SPSS version 13 was used (SPSS Inc., Chicago, Illinois).
A total of 51 patients were enrolled in the study. Of those, 48 patients were randomized and completed 1 year follow-up. Patients’ baseline clinical characteristics are summarized in Table 1.There were no statistically significant differences between the 2 groups in terms of demographic data, etiology of HF, and background medications at baseline and over the course of follow-up period.
Effects on LVEF, volumes, and mass index by CMR
The LVs were dilated at baseline in both groups, indicating severe LV adverse remodeling. There were no significant differences of all baseline CMR parameters between the 2 groups (Table 2).The combination group had significant improvement of LVEF at 26 weeks (Table 3)and further improved at 52 weeks (Table 4).Furthermore, there were significant reductions in LVEDVI, LVESVI, and LV mass index after 1 year of treatment in the combination group (all p < 0.05) (Table 4). In fact, the reduction of volume indexes occurred as early as 26 weeks. In contrast, all these parameters were unchanged in the control group (Tables 3 and 4). The intergroup differences for changes in LVEF, LVEDVI, and LVESVI at 52 weeks were statistically significant by repeated measure ANOVA for trends (Fig. 2).
LV long-axis function by TDI
The Sm and basal peak systolic strain increased significantly, and there was a marked reduction in E/Em only in the combination group (Tables 3 and 4). These echocardiographic parameters were unchanged in the control group.
The CVIB increased significantly in the combination group but not in the control group. This suggested a reduction in myocardial fibrosis after 52 weeks of combination therapy (Table 4).
New York Heart Association (NYHA) functional class, quality of life, and 6-min walking test
The quality of life score was improved similarly in both groups. None of the patients had deterioration in NYHA functional class or significant improvement in the 6-min hall-walk test at the end of the study (Table 5).Two patients in the combination group were hospitalized for congestive HF compared with 4 in the placebo group (p = NS), although this study was not powered to investigate the effect on hospital stays or mortality.
Safety and tolerability of candesartan/spironolactone
There were 3 patients prematurely withdrawn from the study; 1 patient in the combination group and 1 patient in the control group had sustained hypotension. After 1 year, the mean systolic blood pressure decreased in the combination group (127 ± 3 mm Hg to 117 ± 3 mm Hg, p < 0.05). One patient in the combination group had hyperkalemia (5.8 mmol/l) leading to withdrawal from study. The mean serum creatinine level increased in the combination group (from 114 ± 23 μmol/l to 138 ± 69 μmol/l, p < 0.05), although none of the patients reached a level that necessitated withdrawal from the study.
This clinical study addresses the beneficial role of dual blockade of angiotensin II and aldosterone receptors on the reverse remodeling process in chronic mild-to-moderate systolic HF. We were able to demonstrate that combination therapy of candesartan and spironolactone produced substantial reverse remodeling in HF, despite maximal medical therapy including beta-blocker medications and in predominantly NYHA functional class II patients. Dual blockade achieves a significant reduction in LV size, improvement in overall systolic function, and reversal of LV hypertrophy. The improvement of CVIB might also reflect the reduction of myocardial fibrosis. These beneficial effects are apparent early after initiation of therapy, and further improvement occurred throughout the entire 1-year treatment period irrespective of HF etiology. Our study provides a rationale for the use of this combination therapy in selected HF patients. The potential role of dual blockade in further reduction of hospital stays for HF or mortality needs to be addressed by a larger clinical trial.
We postulate that one of the cardioprotective mechanisms of dual blockade is possibly through reduction of myocardial fibrosis and reverse remodeling (10). The combination therapy of spironolactone and candesartan resulted in reduction of LV volumes and myocardial fibrosis in an animal study (11). In this study, CVIB assessment showed improvement after combination therapy similar to that observed in hypertensive HF and in keeping with the documented antifibrotic action of aldosterone antagonists seen experimentally (12).
Another potential beneficial mechanism of combination therapy is more complete inhibition of the renin angiotensin aldosterone system (RAAS). Nonsustained suppression of the RAAS is seen after chronic therapy of beta-blocker (13), ACE inhibitor medications, or ARBs, suggesting that rennin and aldosterone “breakthrough” do occur after the chronic use of current standard anti-HF therapy. Thus, it can be advised that an aldosterone antagonist should be added when an ARB is used alone in treating a patient with systolic HF, because the impact on reverse remodeling is considerable. However, 1 patient (4%) who was taking a combination of an ARB, beta-blocker, and spironolactone developed significant hyperkalemia. Close monitoring of potassium level, careful patient selection, and titration of medication dosage are recommended.
The dose of candesartan used in this study was relatively modest compared with the higher doses used in the CHARM (Candesartan in Heart Failure–Assessment of Mortality and morbidity) trial (14). But in the CHARM study, only about two-thirds of the study population tolerated the target dose of 32 mg because of hypotension and only 24% of patients in the CHARM-Alternative arm were taking spironolactone (15). Lower-dose candesartan has been proven to be efficacious (16). Candesartan 8mg was considered a reasonable therapeutic dose, and we were concerned about the potential risk of hyperkalemia when combining spironolactone with higher doses of candesartan. Of course, our study cannot answer whether spironolactone added to an ARB in larger doses will produce a different result with respect to LV reverse remodeling or the incidence of adverse events. We have provided evidence of a significant beneficial effect on LV reverse remodeling with a relatively low adverse event rate with a low-dose combination therapy in carefully selected patients. The optimal dosage of combining candesartan with spironolactone awaits future studies.
The sample size was relatively small but had adequate power, because we used CMR for volumetric measurements to assess LV reverse remodeling (17). There was a trend in the combination group for lower EF, younger age, less hypertension, coronary artery disease, myocardial infarction, revascularization, more nonischemic etiology, and a 20% less use of diuretic and statin in the combination group, but all these parameters did not reach statistical significance (p > 0.05). The positive results achieved in this study were not likely due to fluid overload at baseline, because diuretic dosage was optimized before recruitment and there was no indication to alter diuretic dosages during the study. Furthermore, the changes in LV dimensions and LVEF were used to assess remodeling in this study. Although LVEF might be influenced by transient loading conditions, the dimensional and volumetric changes probably reflect structural modifications occurring in the myocardium. The role of statin therapy on LV remodeling in HF patients is still controversial.
The present study demonstrates that in patients with chronic mild-to-moderate systolic HF, adding spironolactone to an ARB produces incremental beneficial effects on LV reverse remodeling and LV function. Further larger clinical studies are needed to determine whether spironolactone in conjunction with an ARB can reduce mortality or hospital stays for congestive HF.
The authors would like to acknowledge the efforts of all CMR, echocardiographic, and research staff who participated in the study.
Takeda (Hong Kong) Ltd. donated 25% of the study costs to support this study.
- Abbreviations and Acronyms
- angiotensin-converting enzyme
- angiotensin II receptor blocker
- cardiac magnetic resonance imaging
- cyclic variation of integrated backscatter
- heart failure
- left ventricle/ventricular
- left ventricular end-diastolic volume index
- left ventricular ejection fraction
- left ventricular end-systolic volume index
- tissue Doppler imaging
- Received January 24, 2007.
- Revision received March 9, 2007.
- Accepted March 12, 2007.
- American College of Cardiology Foundation
- Matsumura K.,
- Fujji K.,
- Oniki H.,
- et al.
- Sato A.,
- Saruta T.
- Cohn J.N.,
- Anand I.S.,
- Latini R.,
- et al.,
- Valsartan Heart Failure Trial Investigators
- Yu C.M.,
- Fung J.W.H.,
- Zhang Q.,
- et al.
- Nagueh S.F.,
- Middleton K.J.,
- Kopelen H.A.,
- Zoghbi W.A.,
- Quinones M.A.
- Mottram P.M.,
- Haluska B.,
- Leano R.,
- et al.
- Fraccarollo D.,
- Galuppo R.,
- Schmidt I.,
- et al.
- Young J.B.,
- Dunlap M.E.,
- Pfeffer M.A.,
- et al.
- Yusuf S.,
- Pfeffer M.A.,
- Swedberg K.,
- et al.,
- CHARM Investigators and Committees