Abstract
Studies have shown that dual therapy with angiotensin-converting enzyme inhibitors (ACEI) and either angiotensin II receptor blockers or aldosterone receptor antagonists is more effective in reducing proteinuria than either agent used alone. The questions that remain are as follows: (1) Which of these agents should be used as dual therapy with the ACEI? (2) Does a higher level of blockade of the renin-angiotensin-aldosterone system with triple therapy offer an advantage over dual blockade? A 3-mo randomized, double-blind, placebo-controlled study was performed in 41 patients with proteinuria >1.5 g/d. Four treatment groups were compared: (1) Ramipril + spironolactone placebo + irbesartan placebo, (2) ramipril + irbesartan + spironolactone placebo, (3) ramipril + irbesartan placebo + spironolactone, and (4) ramipril + irbesartan + spironolactone. The percentage change in protein excretion differed according to treatment arm (ANOVA: F3,35 = 8.6, P < 0.001). Pair-wise comparison showed that greater reduction in protein excretion occurred in treatment regimens that incorporated spironolactone. The reduction in proteinuria at 3 mo was as follows: Group 1, 1.4%; group 2, 15.7%; group 3, 42.0%; and group 4, 48.2%. The reduction in proteinuria among patients who were taking spironolactone-containing regimens was sustained at 6 and 12 mo. This study suggests that aldosterone receptor blockade offers a valuable adjuvant treatment when used with ACEI therapy for the reduction of proteinuria. Results suggest no advantage of triple blockade over dual blockade of the renin-angiotensin-aldosterone system to reduce proteinuria.
The number of patients who present with ESRD that requires hemodialysis or transplantation is increasing, highlighting the need to find strategies that slow progression of kidney disease. The need for these strategies is even more critical in countries in which dialysis in not a treatment option. Ideally, new treatment strategies should be readily accessible and not financially prohibitive. The importance of proteinuria as a significant risk factor for ESRD is increasingly recognized (1), and treatment that is targeted at reducing proteinuria has been shown to reduce progression of renal disease (2–7).
Numerous clinical trials have established that interruption of the renin-angiotensin-aldosterone system (RAAS) cascade with angiotensin-converting enzyme inhibitors (ACEI) (2,3) or angiotensin receptor blockers (ARB) (4) is beneficial in slowing progression of renal disease. However, the use of these agents has limitations, with significant numbers of treated patients still demonstrating progressive renal disease (2,5).
Several studies have raised the possibility that aldosterone has a role in mediating renal damage (8,9). Although it may be anticipated that using ACE or ARB would prevent any aldosterone-mediated effects, this does not seem to be the case. In 2001, we reported an uncontrolled study in which we treated eight patients who had chronic kidney disease and persistent proteinuria with the aldosterone receptor antagonist spironolactone in addition to ACE therapy and demonstrated a 54% reduction in proteinuria after 4 wk of treatment (10). Since then, others have confirmed the effectiveness of aldosterone receptor blockade in reducing proteinuria (11–16). Studies have shown that dual therapy with ACE and ARB (6) or ACE with aldosterone receptor antagonists (12,15) is more effective in reducing proteinuria than either agent used alone. With regard to the lowering of proteinuria, we wished to address the following questions: (1) Is angiotensin receptor blockade or aldosterone receptor antagonism more effective in combination with ACE inhibition? (2) Does a higher level of blockade of the RAAS with triple therapy offer an advantage over dual blockade?
Materials and Methods
Between January 2002 and September 2004, patients who were seen in the Nephrology Department of the Royal Melbourne Hospital were screened for the study. Inclusion criteria were (1) age 18 to 75 yr, (2) 24-h urinary protein excretion that exceeded 1.5 g/24 h on at least two occasions at least 3 mo apart, (3) serum creatinine level ≤200 μmol/L with <20% variability in the preceding 3 mo, and (4) treatment with an ACE for at least 6 mo. Exclusion criteria were (1) diastolic BP (DBP) ≥115 mmHg or systolic BP (SBP) ≥220 mmHg; (2) serum potassium level ≥5.0 mm/L; (3) serum bicarbonate level ≤20 mmol/L; (4) acute myocardial infarction or stroke in the previous 6 mo; (5) treatment with corticosteroids, nonsteroidal anti-inflammatory drugs, or immunosuppressant drugs; and (6) evidence or suspicion of renovascular disease, obstructive uropathy, collagen disease, cancer, drug or alcohol abuse, pregnancy, breastfeeding, or ineffective contraception.
Study Design
The study was a single-center, double-blind, placebo-controlled, randomized intervention trial that comprised a run-in phase, a compliance phase, and a treatment phase (Figure 1). Three months after randomization, the patient codes were opened. All patients remained on the allocated treatment until 6 mo and were given the option to start spironolactone. Treatment was continued on an open-label basis until 12 mo. The Melbourne Health Human Ethics committee approved the protocol, and all participants gave written informed consent. This study was investigator initiated and entirely internally funded.
Study design. ACEI, angiotensin-converting enzyme inhibitor.
The randomization process was undertaken by the Royal Melbourne Hospital Pharmacy Department. Treatment assignment was by simple randomization. Clinical trial pharmacists, who were otherwise not involved in the clinical trial, performed treatment allocation. After screening, all eligible patients entered a run-in phase to ensure that they were treated similarly before the commencement of the trial. During this phase, all participants were changed to the oral ACE ramipril 10 mg/d, and DBP was treated to <90 mmHg. Once participants had been on ramipril 10 mg/d for at least 1 mo and achieved the target DBP, they entered a 3-wk compliance phase to ensure that they took medications prescribed.
During the compliance phase, the ramipril dose was reduced from 10 to 5 mg/d, and participants were given two placebo medications to be taken with their morning ramipril dose. They also were given dietary advice about reduced sodium intake, reduced potassium intake, and protein restriction (0.6 to 0.8 g protein/kg body wt per 24 h). Compliance was assessed by pill count.
Participants then were randomized to one of four treatment groups for 12 wk: (1) group 1, ramipril, irbesartan placebo, and spironolactone placebo; (2) group 2, ramipril, irbesartan, and spironolactone placebo; (3) group 3, ramipril, irbesartan placebo, and spironolactone; or (4) group 4, ramipril, irbesartan, and spironolactone. The commencing dosages were ramipril 5 mg/d, irbesartan 150 mg/d, and spironolactone 25 mg/d; these doses were maintained for the duration of the study unless hyperkalemia occurred. The target DBP of < 90 mmHg was maintained during the compliance and treatment phases with BP-lowering agents other than ACEI, ARB, spironolactone, or nondihydropyridine calcium antagonists. When serum potassium rose to >6.0 mmol/L, diuretic therapy was commenced or dosage was increased, the dosage of ramipril was reduced, or the dosage of spironolactone or spironolactone placebo was reduced to 25 mg alternate days.
Twenty-four-hour urinary protein excretion and Cockcroft-Gault calculated creatinine clearance were measured at the beginning and the end of the compliance phase, at the end of the 12-wk treatment phase, and at 6 and 12 mo. BP; body weight; and serum potassium, bicarbonate, urea, and creatinine levels were measured at the beginning and the end of the compliance phase and every 4 wk during the treatment phase, then every 3 mo thereafter. Serum potassium was measured 1 wk after the treatment phase was started.
Statistical Analyses
The primary study end point was the between-group difference in percentage reduction in 24-h urinary protein excretion after 3 mo of treatment. Secondary end points included (1) between-group difference at 6 mo, (2) percentage reduction of 24-h urine protein excretion at 3 and 6 mo separately for each group, and (3) changes in BP and Cockcroft-Gault calculated creatinine clearance. Safety data that were collected included weight and serum potassium, bicarbonate, urea, and creatinine. A post hoc analysis included comparison in the percentage reduction in protein excretion at 6 and 12 mo among patients who received spironolactone.
On the basis of the results of our pilot study (10), in which spironolactone resulted in an average reduction in 24-h urinary protein excretion of 54% after 1 mo of treatment, we estimated that nine participants were required in each treatment group. This sample size was calculated to achieve 80% power to detect a 35% difference in 24-h urinary protein excretion as a result of the addition of either spironolactone or irbesartan to ramipril compared with ramipril alone, at a 5% α level.
All analyses at 3 and 6 mo were based on the intention-to-treat principle. Analyses at 12 mo were according to treatment received. Data from all randomly assigned participants were included, with the exception of one participant who ceased treatment within the first week of randomization on whom no follow-up data were available. ANOVA was used for between-group comparison at each time point (3 and 6 mo). When ANOVA provided evidence of between-group difference, pair-wise comparisons were made using the two-sample t test with Bonferroni-adjusted P values. Creatinine clearance values were log-transformed. Within each group, the change in protein excretion between baseline and 3 mo and between baseline and 6 mo was tested using a one-sample t test. Baseline characteristics were compared using ANOVA (continuous data) or the χ2 test of homogeneity (gender). Statistical analysis was performed using Stata 7.0 (17).
Results
Compliance Phase
Protein excretion remained at a stable level overall during the 3-wk compliance phase. The mean change overall was −2.04% (95% confidence interval −10.06 to 5.98%). No significant changes were seen in BP, body weight, serum electrolytes, or renal function during the compliance phase. Compliance as assessed by pill count was 100% for all participants.
Baseline Characteristics
Baseline characteristics were similar across all treatment arms (Table 1).
Baseline characteristics at the start of the treatment phasea
Treatment Effects
24-H Urinary Protein Excretion.
The percentage change in 24-h urinary protein excretion at 3 and 6 mo is shown in Figure 2 and Table 2.
Percentage change in 24-h urine protein excretion compared with baseline at 3 and 6 mo by treatment group.
Effect of treatment on BP, 24-h urinary creatinine clearance, proteinuria, and safety dataa
Between-Group Analysis.
At 3 mo, the percentage change in protein excretion at 3 mo differed according to treatment group. (F3,35 = 8.6, P < 0.001). Pair-wise comparisons showed that the greatest reduction occurred among patients who were given spironolactone. Compared with the ramipril group (group 1), the 24-h urinary protein excretion reduction at the end of the 3-mo treatment was greater in the ramipril and spironolactone group (group 3; P = 0.004) and in the ramipril, irbesartan, and spironolactone group (group 4; P < 0.001) but not in the ramipril and irbesartan group (group 2; P = 1.000). Triple therapy offered no advantage to using dual therapy with ramipril and spironolactone, resulting in only an additional 6% reduction in protein excretion. Alternatively, using triple therapy provided an additional 32% change in the reduction of proteinuria compared with the dual therapy of ramipril and irbesartan. These results suggest that among patients who were on the spironolactone-containing regimen, there was no advantage to using triple therapy.
At 6 mo, again, the percentage change in protein excretion differed according to treatment (F3,35 = 4.94, P = 0.005). Pair-wise comparisons showed that the greatest reduction occurred among patients who were given spironolactone (Table 3).
Paired comparisons of percentage change in protein excretion at 3 and 6 moa
At 12 mo, five of 10 patients in groups 1 and three of 10 patients of group 2 had been started on spironolactone. Analysis of the data was according to treatment received. Of patients who were randomly assigned to groups 1 and 2 and subsequently had spironolactone added, the percentage reduction in 24-h urine protein excretion compared with 6 mo was −42.1 (range −61.2 to −29.0) and −37 (range −47 to −25), respectively. This compares with patients in groups 1 and 2 who continued on the randomized regimen: 7.9 (range −25.1 to 6.3) and 10 (range to −27.0 to 100), respectively. The percentage reduction of 24-h urine protein excretion at 12 mo compared with baseline was sustained for patients who were randomly assigned to the spironolactone-containing regimens: Groups 3 and 4, −44.8 (−73.6 to −16.1) and −43.0 (−62.0 to −24.7), respectively.
Effect of Each Treatment Regimen.
At the end of 3 mo of blinded therapy, the percentage reduction in 24-h urine protein excretion compared with baseline was significantly different from zero for the spironolactone-containing regimens: Group 3 (P = 0.001) and group 4 (P < 0.001). No significant reduction in proteinuria was seen for the non–spironolactone-containing regimens: Group 1 (P = 0.840) and group 2 (P = 0.102). Again, at 6 mo, only the regimens that contained spironolactone significantly changed the percentage reduction in 24-h urine protein excretion (group 1 P = 0.965; group 2 P = 0.337; group 3 P < 0.001; group 4 P < 0.001).
Cockcroft-Gault Calculated Creatinine Clearance.
No evidence of differences in creatinine clearance was seen within or between the groups or between the groups at 3 and 6 mo (Table 2).
BP.
No evidence of differences was seen within the groups or between the groups in change in SBP at 3 and 6 mo. Similarly, the DBP was not different either within the groups or between the groups at 3 mo; however, at 6 mo, there was a difference between the groups with the DBP being higher in the ramipril only group (P = 0.046; Table 2). Target DBP of <90 mmHg was achieved in all but one participant in the ramipril group.
Effect of Diabetic Nephropathy.
Twenty-seven (67%) of 41 patients had diabetic nephropathy (Table 1). There was no evidence of interaction between treatment effects at either 3 or 6 mo and cause of nephropathy (diabetic versus nondiabetic nephropathy; likelihood ratio test: 3 mo χ2 (3) = 1.65, P = 0.649; 6 mo χ2 (3) = 4.50, P = 0.213). Estimated treatment effects were similar in patients with and without diabetes (Table 4).
Mean percentage change in urine protein excretion from baseline to 3 mo and from baseline to 6 mo by treatment group: Diabetic versus nondiabetic nephropathya
Co-Treatment.
The distribution of additional BP-lowering drugs that were used to control BP was similar in all groups. They included diuretics, β blockers, central α agonists, peripheral α antagonists, and dihydropyridine calcium antagonists. An average of 2.2, 1.9, 2.6, and 2.3 additional BP-lowering drugs were required in groups 1 through 4, respectively. The percentage of patients who were taking a thiazide or loop diuretic was not significantly different among groups 1 through 4: 70, 70, 90, and 70, respectively.
Safety Data.
One patient withdrew from the study after <1 wk of active therapy with ramipril, complaining of feeling unwell and light headed. There was no evidence of hypotension and no change in his renal function compared with baseline. Gynecomastia was not observed in any of the groups.
There was no significant difference between treatment groups or within the treatment groups in the change in weight, serum bicarbonate, serum creatinine, serum urea, or serum cholesterol during the 3-, 6-, and 12-mo treatment periods. In participants with diabetes, no difference was seen for the change in HbA1c during the 3- and 12-mo treatment periods between or within the treatment groups.
The serum potassium was different at 3 mo between the groups (Table 2). The serum potassium was approximately 0.5 mmol/L higher in the spironolactone-containing regimens. A serum potassium of ≥6.0 mmol/L was seen in one patient in group 3 (ramipril + spironolactone) and occurred at 2 mo and in two patients in group 4, one occurring at 3 mo and one occurring at 6 mo. The underlying renal disease was diabetic nephropathy, IgA nephropathy, and focal sclerosis and segmental hyalinosis, respectively. All responded to a reduction in ramipril dose and the introduction or increased dose of a diuretic. Serum bicarbonate decreased to <20 mmol/L in all three patients.
Discussion
It is generally accepted that lowering protein excretion is a desirable treatment goal in chronic proteinuric kidney disease (1,7). Treatment with drugs that interfere with the RAAS slows the course of chronic diabetic and nondiabetic disease (2,4,5). The renoprotective efficacy of these blocking agents may be related to their ability to reduce protein excretion. It was demonstrated recently that for each halving of proteinuria level between baseline and 12 mo with treatment, risk for end-stage renal failure was reduced by more than half (7). Our study indicates a potentially important role for spironolactone in reducing proteinuria. ACEI initially causes a decrease in aldosterone concentration, but with prolonged use, this suppression is not sustained (18). This may be relevant to the dramatic effects seen in our study, in which all patients had already been taking ACEI for >6 mo and circulating aldosterone levels were presumably not suppressed.
We have shown that the addition of the aldosterone receptor antagonist spironolactone to the ACE ramipril resulted in a 42.0% reduction in 24-h urinary protein excretion at 3 mo in patients with proteinuric chronic renal disease. The reduction in proteinuria was sustained up to 12 mo. There was no advantage to using triple blockade of the RAAS with ramipril, irbesartan, and spironolactone compared with dual therapy with ramipril and spironolactone. Triple therapy did offer an advantage to dual therapy with ramipril and irbesartan, suggesting that spironolactone targets a different mediator of disease. The dosages of ramipril, irbesartan, and spironolactone in this study were chosen to minimize the risk for hyperkalemia in patients who were randomly assigned to triple therapy and are not maximal. Our study does not address whether maximal doses of ACEI and angiotensin receptor antagonist would be as effective in lowering proteinuria as spironolactone-containing regimens; neither does our study address whether a higher dosage of spironolactone would result in an even greater fall in proteinuria.
The effect does not seem to be related directly to BP changes; similarly, we could find no evidence that this related directly to reduction in creatinine clearance. Recently, there has been evidence to suggest that diuretics and low-salt diet may potentiate the effect of renin-aldosterone system–blocking agents (19,20). More than 70% of patients in every group were taking either a loop diuretic or thiazide diuretic, so the effect of spironolactone in reducing proteinuria is over and above that provided by concurrent use of diuretics.
Hyperkalemia is a potentially serious complication of combining spironolactone with ACE (21). The patients in our study were screened carefully for risk for developing hyperkalemia on combined treatment. All patients in this study had serum potassium measured 1 wk after starting the treatment phase and every 3 mo thereafter. Although hyperkalemia was not a clinical problem in our study, we do not advocate the use of combination treatment without such careful screening and ongoing monitoring of candidate patients.
The COOPERATE study demonstrated that combination therapy with ACE and ARB was more effective in minimizing proteinuria and slowing progression of renal disease than either agent alone (6). Recent studies have demonstrated that combination therapy with ARB is more effective than the maximal recommended dosage of ACEI in reducing proteinuria in both patients both with and without diabetes (22,23). The effect of the combination therapy in our study was a −15.7% change in proteinuria. This is quantitatively comparable to that seen in other studies of combination therapy, in which changes in proteinuria ranging from 0 to −40% have been reported in randomized trials of adding angiotensin receptor antagonists to ACEI (24–27). However, it does not reach statistical significance because our study was specifically powered to demonstrate a >35% or greater difference. Adding spironolactone to ramipril was more effective than adding irbesartan, and, indeed, because reduction in ramipril dosage from 10 to 5 mg (Table 1) did not significantly alter protein excretion, it could be argued that adding spironolactone likely is more effective than increasing ramipril above the maximum recommended dose.
This study has reinforced the importance of aldosterone as a mediator of renal disease. The use of spironolactone with an ACE alone or in combination with an ARB resulted in a significant reduction in proteinuria. There was no advantage to using triple blockade of the renin-angiotensin aldosterone system over dual blockade with spironolactone and ACEI. We have demonstrated that adding spironolactone to ACE therapy is an effective tool in the effort to lower proteinuria. However, proteinuria reduction remains a surrogate marker for long-term renal survival, and a larger study with longer follow-up is required to demonstrate whether the results shown in our study translate to improved renal survival in patients with persistent proteinuria.
Acknowledgments
This study was presented in abstract form at the World Congress of Nephrology meeting, Singapore, June 26 to 30, 2005.
We gratefully acknowledge the provision of ramipril at no cost by Sonofi-Aventis and the cooperation of the Royal Melbourne Hospital Department of Nephrology clinic staff in this study.
Footnotes
Published online ahead of print. Publication date available at www.cjasn.org.
- Received September 15, 2005.
- Accepted January 3, 2006.
- Copyright © 2006 by the American Society of Nephrology
References
In this issue
Jump to section
More in this TOC Section
Cited By...
- ACE inhibitor and ARB therapy: Practical recommendations
- Aldosterone Dysregulation With Aging Predicts Renal Vascular Function and Cardiovascular Risk
- Aldosterone Breakthrough during Angiotensin Receptor Blocker Use: More Questions than Answers?
- A Randomized, Double-Blind, Placebo-Controlled Trial of Spironolactone on Carotid Intima-Media Thickness in Nondiabetic Hemodialysis Patients
- Aldosterone: Role in Edematous Disorders, Hypertension, Chronic Renal Failure, and Metabolic Syndrome
- Potassium Homeostasis and Renin-Angiotensin-Aldosterone System Inhibitors
- Addition of Angiotensin Receptor Blockade or Mineralocorticoid Antagonism to Maximal Angiotensin-Converting Enzyme Inhibition in Diabetic Nephropathy
- Aldosterone Antagonists for Preventing the Progression of Chronic Kidney Disease: A Systematic Review and Meta-analysis
- Effects of Additive Therapy With Spironolactone on Proteinuria in Diabetic Patients Already on ACE Inhibitor or ARB Therapy: Results of a Randomized, Placebo-Controlled, Double-Blind, Crossover Trial
- Management of Hypertension in Diabetic Patients With Chronic Kidney Disease