Summary
Background and objectives
Inhibition of the renin-angiotensin-aldosterone system decreases proteinuria and slows estimated GFR decline in patients with type 2 diabetes mellitus with overt nephropathy. Serum aldosterone levels may increase during renin-angiotensin-aldosterone system blockade. The determinants and consequences of this aldosterone breakthrough remain unknown.
Design, setting, participants, & measurements
This study examined the incidence, determinants, and changes associated with aldosterone breakthrough in a posthoc analysis of a randomized study that compared the effect of two angiotensin II receptor blockers in patients with type 2 diabetes mellitus with overt nephropathy.
Results
Of 567 of 860 participants included in this posthoc analysis, 28% of participants developed aldosterone breakthrough, which was defined by an increase greater than 10% over baseline values of serum aldosterone levels after 1 year of angiotensin II receptor blocker treatment. Factors independently associated with aldosterone breakthrough at 1 year were lower serum aldosterone and potassium levels at baseline, higher decreases in sodium intake, systolic BP, and estimated GFR from baseline to 1 year, and use of losartan versus telmisartan. Aldosterone breakthrough at 6 months was not sustained at 1 year in 69% of cases, and it did not predict estimated GFR decrease and proteinuria increase between 6 months and 1 year.
Conclusions
Aldosterone breakthrough is a frequent event 1 year after initiating renin-angiotensin-aldosterone system blockade, particularly in participants exposed to intensive lowering of BP with sodium depletion and short-acting angiotensin II receptor blockers. Short-term serum aldosterone level increases at 6 months are not associated with negative kidney outcomes between 6 months and 1 year.
Introduction
Inhibition of the renin-angiotensin-aldosterone system (RAAS), either through angiotensin-converting enzyme inhibitors (ACEIs) in patients with diabetes type 1 (1) and nondiabetics (2,3) or angiotensin II receptor blockers (ARBs) in patients with type 2 diabetes mellitus (T2DM) (4,5), decreases proteinuria and slows down the decline in GFR in patients with proteinuria. Indeed, residual proteinuria is a predictor of progression to renal failure (6). However, despite RAAS blockade, many patients display high residual proteinuria levels and develop ESRD (7). It is possible that RAAS blockade strategies may be suboptimal, because serum aldosterone levels may increase during RAAS blockade with all ACEIs (8) or ARBs (9) tested so far. The incidence of this aldosterone breakthrough is up to 53% after 1 year of RAAS blockade, depending on the definition used (10). Aldosterone breakthrough is usually defined by the increase in serum aldosterone levels compared with baseline levels before RAAS blockade and ranges from 40% to 53% at 1 year (10). Aldosterone breakthrough seems to be independent of the RAAS blocker dosage (11) and does not differ with ACEI or ARB treatment (12).
Aldosterone breakthrough may be associated with cardiovascular and renal morbidity. Indeed, it may reverse the beneficial effects of an ACEI on left ventricular hypertrophy (13), and the improvement in functional capacity noted in patients with congestive heart failure treated with ACEI is decreased when aldosterone breakthrough is present (14). High serum aldosterone levels are an independent risk factor for in-hospital cardiovascular/renal morbidity or mortality after myocardial infarction (15). In patients with diabetic or IgA nephropathies, aldosterone breakthrough is associated with higher levels of proteinuria levels (8,12), and progression to renal failure may be more rapid in diabetic patients with aldosterone breakthrough (9). Furthermore, serum aldosterone has been reported to induce inflammation and fibrosis in experimental animals (16) and therefore, could accelerate renal damage.
The aim of this posthoc analysis was to establish the incidence, determinants, and changes associated with aldosterone breakthrough after ARB treatment in participants included in the AMADEO study (losArtan versus telMisArtan in hypertensive type-2 DiabEtic patients with Overt nephropathy) (17).
Materials and Methods
Study Design and Patient Population
A total of 860 participants with T2DM with first morning spot urinary albumin creatinine ratio (UACR)≥700 mg/g was included in AMADEO, a prospective, randomized, double-blind, controlled, multicenter, parallel-group trial that compared the effects of losartan versus telmisartan on the UACR after 1 year of treatment (17). Serum aldosterone levels were measured at inclusion, 6 months, and study end (1 year). In this posthoc analysis, we excluded patients with missing serum aldosterone measurements, because either they did not complete the 1-year trial because of events or one serum aldosterone value was missing during follow-up.
After a 4-week placebo washout period, with discontinuation of any ARB, ACEI, or direct vasodilator therapy, eligible participants were randomized to one time daily losartan or telmisartan. During the first 2 weeks, the dose was 50 mg for losartan and 40 mg for telmisartan. For the remaining 50 weeks, the doses of losartan and telmisartan were 100 and 80 mg, respectively. Additional antihypertensive medication (excluding other ARBs, ACEIs, or aldosterone inhibitors) could be administered after forced titration to achieve BP targets (<130/80 mmHg).
Data for the Posthoc Analyses
Data on age, sex, body mass index, seated trough systolic BP (SBP) and diastolic BP (DBP), UACR, urinary sodium creatinine ratio (UNaCR), glycosylated hemoglobin (HbA1c), estimated GFR (eGFR) with simplified Modification of Diet in Renal Disease formula, and additional antihypertensive treatments were available at baseline, 6 months, and 1 year. The 24-hour urinary sodium excretion, a good marker of daily sodium intake, was extrapolated from UNaCR in morning urine spots as previously shown using Mann’s formula (18).
Blood was collected between 7:00 and 10:00 a.m. after the patient had rested in the seated position for 30 minutes to determine serum aldosterone levels at baseline, 6 months, and 1 year. Centrifugation was performed within 30 minutes, and 1-ml plasma aliquots were stored at −80°C. Serum aldosterone levels were measured with a commercially available radioimmunoassay kit (ALDO-RIACT; Schering AG, Berlin, Germany). The normal serum aldosterone range was considered to be 42–201 pg/ml.
Three groups of aldosterone level changes were defined: (1) participants with a true aldosterone breakthrough (defined as an increase of serum aldosterone levels greater than 10% over baseline values to take into account the upper limit of the within- and between-subject coefficients of variation of the assay) (19,20), (2) participants with the expected decrease of serum aldosterone levels greater than 10% below baseline values, and (3) participants with nonclinically significant fluctuation of serum aldosterone levels.
To assess whether aldosterone breakthrough predicted changes in eGFR and albuminuria between 6 months and 1 year, we looked at aldosterone breakthrough at 6 months and analyzed subsequent changes in renal function from 6 months to 1 year.
Statistical Analyses
Normally distributed variables were expressed as means ± SD, and other variables were log-transformed to obtain normal distribution and reported as geometric mean ×/÷ SD. Changes of log-transformed quantitative values over time were expressed as geometric mean percent ×/÷ SD (21). Normally distributed variables were compared by one-way ANOVA. Qualitative values were compared using the chi-squared test.
To study the determinants of aldosterone breakthrough, we first analyzed baseline variables associated with baseline serum aldosterone levels by univariate and multivariate regression analyses. Second, we searched for the baseline and follow-up variables associated with aldosterone breakthrough at 1 year as a binary variable (comparing participants with significant increase or decrease of serum aldosterone levels) or a continuous variable (absolute variation of serum aldosterone between baseline and 1 year) by univariate regression analyses. All the relevant clinical and biologic variables and additional antihypertensive treatments were then included in a multivariate logistic regression analysis with nested models. The first model (M1) included baseline variables, whereas the second model (M2) included both M1 and changes observed in these variables from baseline to 1 year to estimate the adjusted odds ratios and 95% confidence interval for aldosterone breakthrough at 1 year.
Because our definition of aldosterone breakthrough excluded patients with nonsignificant serum aldosterone changes from the statistical analysis, the results were confirmed in the total study population by an analysis of serum aldosterone levels as a continuous variable using a general regression model.
Because Na intake is a strong determinant of serum aldosterone production, additional analyses were performed in participants with increased or decreased sodium intake from baseline to 1 year. After testing for interactions between characteristics and aldosterone breakthrough in the two subgroups of patients with increased or decreased Na intake during follow-up, we performed a multivariate analysis with all the significant interactions in the total study population.
Because of the colinearity between SBP and DBP, only SBP was included in the multivariate analyses, because it was found to have a more statistically significant association with aldosterone breakthrough in univariate analysis.
To address the possibility of regression to the mean, we considered different statistical adjustments for baseline as well as average serum aldosterone levels during follow-up as previously described (22).
Analyses were performed using SAS software v9.1 (SAS institute, Cary, NC). All P values were two-tailed, and P values<0.05 were considered statistically significant.
Results
Baseline Characteristics
Of 860 participants with T2DM randomized in AMADEO, 567 participants were included in this posthoc analysis, because they had completed the 1-year study and had morning serum aldosterone values available at baseline, 6 months, and 1 year. The baseline characteristics of the included and excluded participants are reported in Table 1; 173 participants did not complete the trial (death, renal replacement therapy, or lost to follow-up), and 121 participants had missing values of serum aldosterone during follow-up, including 14 patients with missing values at 6 months.
Table 1.
Baseline characteristics of the AMADEO study population: comparison of participants included and excluded from the posthoc analysis (because they did not complete the trial or have serum aldosterone values at baseline, 6 months, and/or 1 year)
| Characteristics | Participants Included (n=567) | Participants Excluded (n=293) | P Value |
|---|---|---|---|
| Age (yr) mean ± SD | 60.3±9 | 60.1±9 | 0.70 |
| Men n (%) | 364 (64) | 171 (58) | 0.10 |
| Ethnicity n (%) | |||
| Caucasian | 256 (45) | 149 (51) | <0.001 |
| Black | 49 (9) | 53 (18) | <0.001 |
| Asian | 262 (46) | 91 (31)a | <0.001 |
| Body mass index (kg/m2) mean ± SD | 30±6 | 31±7 | 0.10 |
| Smoking status n (%) | |||
| Current | 95 (17) | 39 (13) | 0.14 |
| Previous | 208 (37) | 126 (43) | 0.14 |
| No | 264 (47) | 127 (43) | 0.14 |
| Systolic BP (mmHg) mean ± SD | 143±14 | 145±17 | 0.10 |
| Diastolic BP (mmHg) mean ± SD | 80±9 | 79±10 | 0.50 |
| Serum aldosterone (ng/dl) geometric mean ×/÷ SD | 8.4×/÷2.1 | — | |
| Urinary albumin creatinine ratio (mg/g) geometric mean ×/÷ SD | 1343.3×/÷2.4 | 1507.2×/÷2.9a | 0.007 |
| Na intake (Mann’s formula; g/d) geometric mean ×/÷ SD | 137.4×/÷1.4 | 140.6×/÷1.4 | 0.40 |
| Glycosylated hemoglobin (%) mean ± SD | 7.9±1.3 | 7.9±1.4 | 0.60 |
| Estimated GFR (ml/min per 1.73 m2) geometric mean ×/÷ SD | 45.2×/÷1.6 | 40.5×/÷1.6a | 0.002 |
| Losartan (%) | 51 | 51 | 0.97 |
P<0.05 versus participants included in posthoc analysis.
The statistically significant differences between participants who were included in the posthoc analysis and participants who were excluded were ethnicity, UACR, and eGFR. Indeed, excluded participants were more frequently black and less frequently Asian, and they had higher UACRs and lower eGFRs (Table 1). Baseline serum aldosterone levels were independently and negatively correlated with eGFR, UACR, sodium intake, and serum potassium levels (Supplemental Table 1).
Incidence and Determinants of Aldosterone Breakthrough at 1 Year
Serum aldosterone levels increased in 40% of participants, but only 28% of participants exhibited a true aldosterone breakthrough at 1 year as defined by an increase of serum aldosterone levels greater than 10% over baseline values. After 1 year of ARB treatment, mean changes from baseline were −21×/÷92.5% for serum aldosterone levels, −2.3±19.2 mmHg for SBP, −2.2±11.1 mmHg for DBP, −24×/÷137% for UACR, +5.5×/÷117% for UNaCR, +0.10±1.4% for HbA1c, and −7.3±9.7 ml/min per 1.73 m2 for eGFR in the overall posthoc analysis population.
Univariate analysis showed that factors associated with aldosterone breakthrough at 1 year included higher baseline SBP and UACR and lower baseline serum aldosterone levels (Table 2). The changes in clinical and biologic parameters during follow-up that were significantly associated with aldosterone breakthrough were a greater decrease in SBP, sodium intake, and eGFR and a greater increase of serum potassium from baseline to 1 year (Table 2).
Table 2.
Comparison of baseline characteristics, concomitant medications, and changes during follow-up in participants with a >10% increase versus >10% decrease in serum aldosterone levels from baseline to 1 year (univariate analysis)
| Variables | Serum Aldosterone Level Changes between Baseline and 1 Yr | P Value | |
|---|---|---|---|
| Increase>10% (n=158; 28%) | Decrease>10% (n=320; 56%) | ||
| Baseline characteristics | |||
| Age (yr) mean ± SD | 59.4±9.8 | 60.6±9.2 | 0.19 |
| Men n (%) | 94 (59) | 206 (64) | 0.30 |
| Body mass index (kg/m2) mean ± SD | 29.7±6.4 | 29.6±5.9 | 0.77 |
| Systolic BP (mmHg) mean ± SD | 144.8±14.6 | 141.3±14.4 | 0.01 |
| Diastolic BP (mmHg) mean ± SD | 80.6±9.0 | 79.0±9.0 | 0.07 |
| Serum aldosterone (ng/dl) geometric mean ×/÷ SD | 6.2×/÷1.9 | 10.7×/÷1.9 | <0.001 |
| Glycosylated hemoglobin (%) mean ± SD | 7.9±1.3 | 7.9±1.2 | 0.72 |
| Urinary albumin creatinine ratio (mg/g) geometric mean ×/÷ SD | 1525.4×/÷2.7 | 1281.5×/÷2.2 | 0.01 |
| Na intake (Mann’s formula; mmol/d) geometric mean ×/÷ SD | 140.8×/÷1.4 | 133.5×/÷1.4 | 0.06 |
| Serum potassium (mmol/L) mean ± SD | 4.5±0.5 | 4.5±0.4 | 0.54 |
| Estimated GFR (ml/min per 1.73 m2) geometric mean ×/÷ SD | 43.7×/÷1.5 | 44.7×/÷1.6 | 0.55 |
| Baseline additional antihypertensive treatments | |||
| Diuretics n (%) | 101 (65) | 198 (65) | 0.87 |
| β-blockers n (%) | 83 (53) | 139 (45) | 0.15 |
| Calcium channel blockers n (%) | 115 (73) | 236 (77) | 0.30 |
| Other n (%) | 32 (20) | 58 (19) | 0.74 |
| Angiotensin II receptor blocker treatment from baseline to 1 yr | |||
| Losartan n (%) | 85 (54) | 151 (47) | 0.17 |
| Changes from baseline to 1 yr | |||
| Systolic BP (mmHg) mean ± SD | −6.7±18.6 | −1.4±17.6 | 0.003 |
| Diastolic BP (mmHg) mean ± SD | −4.2±11.4 | −2.1±10.8 | 0.05 |
| Urinary albumin creatinine ratio (mg/g) geometric mean percent ×/÷ SD | −36.4×/÷158.9 | −28.9×/÷149.6 | 0.37 |
| Na intake (Mann’s formula; mmol/d) geometric mean percent ×/÷ SD | −8.3×/÷38.6 | 7.0×/÷33.7 | <0.001 |
| Glycosylated hemoglobin (%) mean ± SD | 0.2±1.4 | 0.1±1.4 | 0.48 |
| Serum potassium (mmol/L) mean ± SD | 0.2±0.5 | 0.1±0.4 | 0.02 |
| Estimated GFR (ml/min per 1.73 m2) geometric mean percent ×/÷ SD | −31.4×/÷29.7 | −14.8×/÷24.0 | <0.001 |
| Diuretics n (%) | +14 (9) | +23 (8) | 0.61 |
| β-blockers n (%) | +10 (6) | +23 (8) | 0.64 |
| Calcium channel blockers n (%) | +13 (8) | +19 (6) | 0.42 |
| Other n (%) | +13 (8) | +24 (8) | 0.88 |
In multivariate analysis, the baseline characteristics independently associated with aldosterone breakthrough at 1 year were higher SBP, lower serum aldosterone, and lower serum potassium levels after adjustment (for age, sex, ethnicity, smoking status, baseline HbA1c, UNaCR, UACR, eGFR, and antihypertensive treatment) (Table 3). After additional adjustment for changes in HbA1c, UACR, UNaCR, serum potassium levels, and additional antihypertensive treatment taken during follow-up (Table 3), the baseline characteristics and follow-up parameters that were statistically significantly associated with aldosterone breakthrough at 1 year were lower serum aldosterone and potassium levels at baseline, higher decreases in SBP, sodium intake, and eGFR from baseline to 1 year, and use of losartan versus telmisartan. When the variation of serum aldosterone levels was analyzed as a continuous variable, statistically significant associations were found with low baseline serum aldosterone and potassium levels, significant decreases in SBP, sodium intake, and eGFR, and significant increases in serum potassium levels and ARB treatment (Table 4). The same results were found with adjustment for average serum aldosterone levels during follow-up instead of baseline serum aldosterone levels.
Table 3.
Predictors of aldosterone breakthrough at 1 year (i.e., serum aldosterone level increase greater than 10% over baseline values) versus serum aldosterone decrease >10% below baseline values using multivariate logistic regression analysis
| Variables | Increase of Serum Aldosterone Levels>10% over Baseline Values (n=158 points) Versus Decrease of Serum Aldosterone Levels>10% below Baseline Values (n=320 points) | |
|---|---|---|
| Model 1a Odds Ratio (95% CI) | Model 2b Odds Ratio (95% CI) | |
| Baseline characteristics | ||
| Serum aldosterone (log mg/dl) | 0.22 (0.15 to 0.33) | 0.19 (0.12 to 0.30) |
| Systolic BP (mmHg) | 1.02 (1.00 to 1.03) | 1.01 (0.98 to 1.03) |
| Serum potassium (mmol/L) | 0.53 (0.32 to 0.86) | 0.47 (0.26 to 0.87) |
| Significant decrease from baseline to 1 yr | ||
| Systolic BP (mmHg) | — | 1.02 (1.00 to 1.04) |
| Na intake (Mann’s formula; log mmol/d) | — | 1.83 (1.24 to 2.7) |
| Serum potassium (mmol/L) | — | 0.77 (0.45 to 1.31) |
| Estimated GFR (log ml/min per 1.73 m2) | — | 8.3 (2.6 to 25.1) |
| Angiotensin II receptor blocker treatment | ||
| Losartan versus telmisartan | — | 1.63 (1.02 to 2.64) |
95% CI, confidence interval.
Adjustment for baseline age, sex, ethnicity, sodium intake, urinary albumin creatinine ratio, estimated GFR, and additional antihypertensive treatment.
Adjustment for urinary albumin creatinine ratio, sodium intake, and additional antihypertensive treatment variations (in addition to the parameters included in model 1).
Table 4.
Baseline and follow-up variables associated with log serum aldosterone level increase between baseline and 1 year using multivariate regression analysis
| Variables | Model 1a | Model 2b | ||
|---|---|---|---|---|
| β ± SD | P Value | β ± SD | P Value | |
| Baseline characteristics | ||||
| Serum aldosterone (log ng/dl) | −0.44±0.04 | <0.001 | −0.40±0.04 | <0.001 |
| Systolic BP (mmHg) | 0.004±0.002 | 0.02 | 0.0005±0.002 | 0.40 |
| Serum potassium (mmol/L) | −0.21±0.06 | <0.001 | −0.14±0.06 | 0.02 |
| Significant decrease from baseline to 1 yr | ||||
| Systolic BP (mmHg) | — | — | 0.005±0.002 | 0.004 |
| Na intake (Mann’s formula; log mmol/d) | 0.33±0.08 | <0.002 | ||
| Serum potassium (mmol/L) | −0.16±0.05 | 0.003 | ||
| Estimated GFR (log ml/min per 1.73 m2) | — | — | 0.50±0.11 | <0.001 |
| Angiotensin II receptor blocker treatment | ||||
| Losartan versus telmisartan | — | — | 0.12±0.05 | 0.01 |
After adjustment for baseline age, sex, ethnicity, smoking status, glycosylated hemoglobin, sodium intake, urinary albumin creatinine ratio, eGFR, and additional antihypertensive treatment.
Adjustment for glycosylated hemoglobin, urinary albumin creatinine ratio, sodium intake, and additional antihypertensive treatment variations (in addition to the parameters included in model 1).
Sensitivity Analysis in Participants Who Decreased Versus Increased Their Sodium Intake during the Study
Aldosterone breakthrough occurred in 93 of 272 participants (34%) who had a decrease of sodium intake and only 65 of 295 participants (22%) who had an increase of sodium intake from baseline to 1 year. The magnitude of Na intake changes during follow-up was the only variable that showed a statistically significant interaction with aldosterone breakthrough in the two subgroups of patients with increased or decreased Na intake during follow-up. The magnitude of Na intake decrease during follow-up (but not Na intake increase) was significantly associated with aldosterone breakthrough (Table 5). The determinants independently associated with aldosterone breakthrough, including this significant interaction with changes of sodium intake during follow-up, were unchanged, except for baseline serum potassium (Table 5). Multivariate analyses in the two subgroups of patients with increased or decreased Na intake during follow-up are shown in Supplemental Table 2, A and B.
Table 5.
Determinants of aldosterone breakthrough at 1 year (i.e., serum aldosterone level increase greater than 10% over baseline values) versus serum aldosterone decrease (>10% below baseline values) associated with aldosterone breakthrough, including the significant interaction with changes of sodium intake during follow-up, using multivariate logistic regression analysis
| Variables | Increase of Serum Aldosterone Levels>10% over Baseline Values (n=158 points) Versus Decrease of Serum Aldosterone Levels>10% below Baseline Values (n=320 points) | |
|---|---|---|
| Model 1a Odds Ratio (95% CI) | Model 2b Odds Ratio (95% CI) | |
| Baseline characteristics | ||
| Serum aldosterone (log mg/dl) | 0.22 (0.17 to 0.34) | 0.23 (0.15 to 0.35) |
| Systolic BP (mmHg) | 1.02 (1.00 to 1.03) | 1.01 (0.99 to 1.03) |
| Serum potassium (mmol/L) | 0.53 (0.32 to 0.86) | 0.64 (0.38 to 1.07) |
| Significant decrease from baseline to 1 yr | ||
| Systolic BP (mmHg) | — | 1.02 (1.00 to 1.03) |
| Serum potassium (mmol/L) | — | 0.78 (0.48 to 1.27) |
| Estimated GFR (log ml/min per 1.73 m2) | — | 6.44 (2.31 to 17.95) |
| Na intake (Mann’s formula; log mmol/d) | ||
| Subgroup with increased Na intake | — | 1.26 (0.12 to 13.3) |
| Subgroup with decreased Na intake | — | 4.45 (1.38 to 14.3) |
| Angiotensin II receptor blocker treatment | ||
| Losartan versus telmisartan | — | 1.60 (1.04 to 2.53) |
95% CI, confidence interval.
After adjustment for baseline age, sex, ethnicity, smoking status, glycosylated hemoglobin, sodium intake, urinary albumin creatinine ratio, estimated GFR, and additional antihypertensive treatment.
With adjustment for glycosylated hemoglobin, urinary albumin creatinine ratio, sodium intake, and additional antihypertensive treatment variations (in addition to the parameters included in model 1).
Changes in Study Parameters between 6 Months and 1 Year Associated with Aldosterone Breakthrough at 6 Months
Aldosterone breakthrough was observed in 163 participants at 6 months, and 112 of these participants (69%) did not present aldosterone breakthrough at 1 year. Multivariate analysis showed that aldosterone breakthrough at 6 months did not predict eGFR decrease and proteinuria increase between 6 months and 1 year (Table 6).
Table 6.
Variables associated with aldosterone breakthrough at 6 months (i.e., serum aldosterone level increase greater than 10% over baseline values) versus serum aldosterone decrease>10% below baseline values using multivariate logistic regression analysis
| Variables | Serum Aldosterone Increase>10% between Baseline and 6 Mo (n=163) Versus Decrease>10% between Baseline and 6 Mo (n=312) | |
|---|---|---|
| Model 1a Odds Ratio (95% CI) | Model 2b Odds Ratio (95% CI)) | |
| Patient characteristics and treatments at 6 mo | ||
| Serum aldosterone (ng/dl) | 4.86 (3.28 to 7.19) | 5.09 (3.34 to 7.77) |
| Systolic BP (mmHg) | 0.99 (0.98 to 1.01) | 1.00 (0.98 to 1.02) |
| Estimated GFR (log ml/min per 1.73 m2) | 1.31 (0.76 to 2.25) | 1.61 (0.89 to 2.89) |
| Urinary albumin creatinine ratio (log mg/g) | 0.98 (0.79 to 1.21) | 1.02 (0.80 to 1.29) |
| Na intake (Mann’s formula; log mmol/d) | 1.16 (0.57 to 2.38) | 1.49 (0.64 to 3.45) |
| Serum potassium (mmol/L) | 1.77 (1.15 to 2.70) | 1.75 (1.10 to 3.11) |
| Diuretics | 0.50 (0.29 to 0.87) | 0.46 (0.25 to 0.83) |
| β-blockers | 0.83 (0.52 to 1.31) | 0.85 (0.52 to 1.39) |
| Calcium channel blockers | 1.18 (0.69 to 2.02) | 0.92 (0.50 to 1.72) |
| Other | 0.81 (0.48 to 1.36) | 0.79 (0.45 to 1.37) |
| Significant decrease after 6 mo | ||
| Systolic BP (mmHg) | — | 0.99 (0.98 to 1.01) |
| Urinary albumin creatinine ratio (log mg/g) | — | 0.83 (0.58 to 1.19) |
| Na intake (Mann’s formula; log mmol/d) | — | 0.91 (0.35 to 2.38) |
| Estimated GFR (log ml/min per 1.73 m2) | — | 0.72 (0.19 to 2.68) |
| Serum potassium (mmol/L) | — | 0.99 (0.57 to 1.71) |
| Changes in antihypertensive treatments after 6 mo | ||
| Diuretics | — | 0.45 (0.10 to 2.16) |
| β-blockers | — | 1.30 (0.32 to 5.25) |
| Ca inhibitors | — | 0.30 (0.09 to 0.98) |
| Other | — | 0.64 (0.21 to 1.99) |
| Angiotensin II receptor blocker treatment | ||
| Losartan versus telmisartan | — | 1.64 (1.03 to 2.62) |
95% CI, confidence interval.
After adjustment for baseline age, sex, and ethnicity.
Model 1 plus changes in study parameters between 6 months and 1 year.
Discussion
Our study provides the largest-scale confirmation so far that aldosterone breakthrough is a frequent event 1 year after initiating RAAS blockade, and it is the first study to provide data suggestive of potential mechanisms behind aldosterone breakthrough.
At baseline, serum aldosterone levels were higher in participants with lower sodium intake and serum potassium levels. Although the diuretic dosage was not reported during the study, lower serum potassium levels suggest higher doses of thiazide or loop diuretics, which may, together with lower sodium intake, stimulate aldosterone synthesis. Indeed, a lower serum potassium level would feed back to decrease serum aldosterone levels in the absence of hypovolemia (23). Sodium depletion may also be the link between high serum aldosterone levels and low UACR and eGFR at baseline, which was previously shown (24–26). Inversely, subsequent aldosterone breakthrough was more frequent if the patient had, at baseline, the opposite profile suggestive of hypervolemia (i.e., lower serum aldosterone levels and higher SBP in multivariate analysis as well as higher UACR in univariate analysis). Serum potassium levels were forced in the multivariate model, despite the absence of association with aldosterone breakthrough in univariate analysis, because of the known physiologic relationship between serum potassium and aldosterone levels. Indeed, high serum potassium levels should favor an increase of aldosterone secretion. However, the respective roles of potassium and diuretic treatments on aldosterone breakthrough remain uncertain, because neither the diuretic dosage nor potassium intake and urine potassium was reported during the study.
The multivariate analysis with follow-up variables showed that aldosterone breakthrough was associated with a greater decrease in sodium intake, SBP, and eGFR over the study period. This result suggests that intensive BP- and proteinuria-lowering strategies with sodium restriction may be a triggering factor for the development of aldosterone breakthrough, which may then counteract part of the beneficial effects of low BP in the long term. When we analyzed the subgroup of participants with a decrease of sodium intake from baseline to 1 year, the main determinant of aldosterone breakthrough was, indeed, a lower sodium intake. Furthermore, in the subgroup of participants who increased their sodium intake during the study, the main determinants of aldosterone breakthrough were lower SBP and eGFR, suggesting again a direct effect of hemodynamic changes.
Serum potassium levels increased during the study after the introduction of ARB and with the decrease of eGFR during follow-up. Higher serum potassium levels should favor an increase of serum aldosterone levels by a physiologic feedback loop. Indeed, the observed increase of serum potassium was statistically significantly associated with an increased risk of aldosterone breakthrough.
Losartan exposed participants to a higher risk of aldosterone breakthrough than telmisartan, an effect that was independent of all the other covariables. Indeed, 57% of subjects with aldosterone breakthrough were taking losartan. Aldosterone secretion is subject to nycthemeral variations, with higher levels during the night that may be further increased by volume depletion during the day using loop diuretics (23). Short-acting RAAS blockers, such as losartan, may not effectively block the peak nocturnal RAAS activity when administered in the morning and therefore, could favor aldosterone breakthrough. This finding might explain why short-acting RAAS blockers may be more efficient in lowering ambulatory BP when administered in the evening (27,28). However, long-acting ARB is not the only and definitive answer to prevent aldosterone breakthrough. Indeed, 43% of subjects with aldosterone breakthrough were taking telmisartan.
We did not observe a faster eGFR decrease between 6 months and 1 year in participants with aldosterone breakthrough at 6 months. In fact, the aldosterone breakthrough status is not definitive at 6 months in this intervention study with ongoing forced titration to achieve BP targets<130/80 mmHg. Long-term study would be required to assess the consequence of sustained aldosterone breakthrough.
The major strength of this study is the well characterized population of participants with T2DM with overt nephropathy who were followed in a prospective, randomized, double-blind, double-dummy, forced titration, multicenter, parallel-group study; this study has the largest number of participants reported so far with centralized measurement of serum aldosterone levels and extensive data collection, including additional antihypertensive treatments, UNaCR, UACR, and serum potassium levels, to adjust for potential confounding factors.
However, our posthoc analysis has several limitations. First, it excluded participants with a missing serum aldosterone measurement because of incomplete follow-up during the first year, mostly because of events but also because of missing values. The baseline characteristics of participants excluded because of incomplete serum aldosterone values differed regarding ethnicity, with fewer black participants, as well as higher UACR and lower eGFR (Table 1). Second, regression to the mean may be mentioned, because lower baseline serum aldosterone levels were associated with subsequent higher incidence of aldosterone breakthrough. However, the appropriate statistical adjustments for baseline as well as average serum aldosterone levels were performed as previously described (22). Serum aldosterone levels were measured at baseline, 6 months, and study end at 1 year. However, we may miss potential day-to-day variations, which may only be partially addressed by the definition of aldosterone breakthrough as a 10% increase over baseline values to take into account the upper limit of the within- and between-subject coefficients of variation of the assay previously reported in the literature. Indeed, these coefficients of variation are fairly high (19). Furthermore, we did not measure urine aldosterone and renin. Third, sodium intake was not assessed but indirectly evaluated with Mann’s formula from spot UNaCR values as previously shown (18). Fourth, diuretic dosages were not reported during the study. However, a diuretic effect may be extrapolated from serum potassium levels, because only thiazide and loop diuretics were allowed during the study, although we did not measure urine potassium and potassium intake. Fifth, no information on whether the ARB was taken in the morning or the evening is available. Because the time of drug intake was not predefined in this study, some participants may have taken the short-acting ARB (losartan) in the evening, thereby decreasing the power of the comparison between the two ARBs if the hypothesis of the need for blocking nighttime RAAS overactivity to reduce aldosterone breakthrough is accepted. Finally, the 1-year study duration did not allow the long-term detrimental consequences of aldosterone breakthrough to be thoroughly investigated.
It can be concluded that aldosterone breakthrough is a frequent event in participants with T2DM with overt nephropathy after the initiation of ARB treatment, and it may be favored by baseline hypervolemia, subsequent greater decrease in SBP, sodium intake, and eGFR during follow-up, and use of a short-acting ARB. Long-term studies are required to test whether prevention of aldosterone breakthrough by cautious dual ACEI/ARB blockade in specific patient populations (29,30) or aldosterone inhibitors (31) would lead to improved nephroprotection.
Disclosures
None.
Supplementary Material
Acknowledgments
Data from the AMADEO study were provided by Boehringer Ingelheim, which allowed us to perform this independent posthoc analysis.
Footnotes
Published online ahead of print. Publication date available at www.cjasn.org.
This article contains supplemental material online at http://cjasn.asnjournals.org/lookup/suppl/doi:10.2215/CJN.06960712/-/DCSupplemental.
See related editorial, “Aldosterone Breakthrough during Angiotensin Receptor Blocker Use: More Questions than Answers?,” on pages 1637–1639.
References
- 1.Lewis EJ, Hunsicker LG, Bain RP, Rohde RD, The Collaborative Study Group : The effect of angiotensin-converting-enzyme inhibition on diabetic nephropathy. N Engl J Med 329: 1456–1462, 1993 [DOI] [PubMed] [Google Scholar]
- 2.Jafar TH, Schmid CH, Landa M, Giatras I, Toto R, Remuzzi G, Maschio G, Brenner BM, Kamper A, Zucchelli P, Becker G, Himmelmann A, Bannister K, Landais P, Shahinfar S, de Jong PE, de Zeeuw D, Lau J, Levey AS: Angiotensin-converting enzyme inhibitors and progression of nondiabetic renal disease. A meta-analysis of patient-level data. Ann Intern Med 135: 73–87, 2001 [DOI] [PubMed] [Google Scholar]
- 3.Jafar TH, Schmid CH, Levey AS: Effect of angiotensin-converting enzyme inhibitors on progression of nondiabetic renal disease. Ann Intern Med 137: 298–299, 2002 [DOI] [PubMed] [Google Scholar]
- 4.Brenner BM, Cooper ME, de Zeeuw D, Keane WF, Mitch WE, Parving HH, Remuzzi G, Snapinn SM, Zhang Z, Shahinfar S, RENAAL Study Investigators : Effects of losartan on renal and cardiovascular outcomes in patients with type 2 diabetes and nephropathy. N Engl J Med 345: 861–869, 2001 [DOI] [PubMed] [Google Scholar]
- 5.Lewis EJ, Hunsicker LG, Clarke WR, Berl T, Pohl MA, Lewis JB, Ritz E, Atkins RC, Rohde R, Raz I, Collaborative Study Group : Renoprotective effect of the angiotensin-receptor antagonist irbesartan in patients with nephropathy due to type 2 diabetes. N Engl J Med 345: 851–860, 2001 [DOI] [PubMed] [Google Scholar]
- 6.de Zeeuw D, Remuzzi G, Parving HH, Keane WF, Zhang Z, Shahinfar S, Snapinn S, Cooper ME, Mitch WE, Brenner BM: Proteinuria, a target for renoprotection in patients with type 2 diabetic nephropathy: Lessons from RENAAL. Kidney Int 65: 2309–2320, 2004 [DOI] [PubMed] [Google Scholar]
- 7.Suissa S, Hutchinson T, Brophy JM, Kezouh A: ACE-inhibitor use and the long-term risk of renal failure in diabetes. Kidney Int 69: 913–919, 2006 [DOI] [PubMed] [Google Scholar]
- 8.Sato A, Hayashi K, Naruse M, Saruta T: Effectiveness of aldosterone blockade in patients with diabetic nephropathy. Hypertension 41: 64–68, 2003 [DOI] [PubMed] [Google Scholar]
- 9.Schjoedt KJ, Andersen S, Rossing P, Tarnow L, Parving HH: Aldosterone escape during blockade of the renin-angiotensin-aldosterone system in diabetic nephropathy is associated with enhanced decline in glomerular filtration rate. Diabetologia 47: 1936–1939, 2004 [DOI] [PubMed] [Google Scholar]
- 10.Bomback AS, Klemmer PJ: The incidence and implications of aldosterone breakthrough. Nat Clin Pract Nephrol 3: 486–492, 2007 [DOI] [PubMed] [Google Scholar]
- 11.MacFadyen RJ, Lee AF, Morton JJ, Pringle SD, Struthers AD: How often are angiotensin II and aldosterone concentrations raised during chronic ACE inhibitor treatment in cardiac failure? Heart 82: 57–61, 1999 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Horita Y, Taura K, Taguchi T, Furusu A, Kohno S: Aldosterone breakthrough during therapy with angiotensin-converting enzyme inhibitors and angiotensin II receptor blockers in proteinuric patients with immunoglobulin A nephropathy. Nephrology (Carlton) 11: 462–466, 2006 [DOI] [PubMed] [Google Scholar]
- 13.Sato A, Saruta T: Aldosterone escape during angiotensin-converting enzyme inhibitor therapy in essential hypertensive patients with left ventricular hypertrophy. J Int Med Res 29: 13–21, 2001 [DOI] [PubMed] [Google Scholar]
- 14.Cicoira M, Zanolla L, Franceschini L, Rossi A, Golia G, Zeni P, Caruso B, Zardini P: Relation of aldosterone “escape” despite angiotensin-converting enzyme inhibitor administration to impaired exercise capacity in chronic congestive heart failure secondary to ischemic or idiopathic dilated cardiomyopathy. Am J Cardiol 89: 403–407, 2002 [DOI] [PubMed] [Google Scholar]
- 15.Beygui F, Montalescot G, Vicaut E, Rouanet S, Van Belle E, Baulac C, Degrandsart A, Dallongeville J, OPERA Investigators : Aldosterone and long-term outcome after myocardial infarction: A substudy of the french nationwide Observatoire sur la Prise en charge hospitalière, l’Evolution à un an et les caRactéristiques de patients présentant un infArctus du myocarde avec ou sans onde Q (OPERA) study. Am Heart J 157: 680–687, 2009 [DOI] [PubMed] [Google Scholar]
- 16.Greene EL, Kren S, Hostetter TH: Role of aldosterone in the remnant kidney model in the rat. J Clin Invest 98: 1063–1068, 1996 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Bakris G, Burgess E, Weir M, Davidai G, Koval S, AMADEO Study Investigators : Telmisartan is more effective than losartan in reducing proteinuria in patients with diabetic nephropathy. Kidney Int 74: 364–369, 2008 [DOI] [PubMed] [Google Scholar]
- 18.Mann SJ, Gerber LM: Estimation of 24-hour sodium excretion from spot urine samples. J Clin Hypertens (Greenwich) 12: 174–180, 2010 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Ricós C, Alvarez V, Cava F, García-Lario JV, Hernández A, Jiménez CV, Minchinela J, Perich C, Simón M: Current databases on biological variation: Pros, cons and progress. Scand J Clin Lab Invest 59: 491–500, 1999 [DOI] [PubMed] [Google Scholar]
- 20.Vasan RS, Demissie S, Kimura M, Cupples LA, Rifai N, White C, Wang TJ, Gardner JP, Cao X, Benjamin EJ, Levy D, Aviv A: Association of leukocyte telomere length with circulating biomarkers of the renin-angiotensin-aldosterone system: The Framingham Heart Study. Circulation 117: 1138–1144, 2008 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Bland JM, Altman DG: The use of transformation when comparing two means. BMJ 312: 1153, 1996 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Cain KC, Diehr P: Testing the null hypothesis in small area analysis. Health Serv Res 27: 267–294, 1992 [PMC free article] [PubMed] [Google Scholar]
- 23.Quinn SJ, Williams GH: Regulation of aldosterone secretion. Annu Rev Physiol 50: 409–426, 1988 [DOI] [PubMed] [Google Scholar]
- 24.Esnault VL, Ekhlas A, Delcroix C, Moutel MG, Nguyen JM: Diuretic and enhanced sodium restriction results in improved antiproteinuric response to RAS blocking agents. J Am Soc Nephrol 16: 474–481, 2005 [DOI] [PubMed] [Google Scholar]
- 25.Esnault VL, Ekhlas A, Nguyen JM, Moranne O: Diuretic uptitration with half dose combined ACEI + ARB better decreases proteinuria than combined ACEI + ARB uptitration. Nephrol Dial Transplant 25: 2218–2224, 2010 [DOI] [PubMed] [Google Scholar]
- 26.Vogt L, Waanders F, Boomsma F, de Zeeuw D, Navis G: Effects of dietary sodium and hydrochlorothiazide on the antiproteinuric efficacy of losartan. J Am Soc Nephrol 19: 999–1007, 2008 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Hermida RC, Calvo C, Ayala DE, López JE: Decrease in urinary albumin excretion associated with the normalization of nocturnal blood pressure in hypertensive subjects. Hypertension 46: 960–968, 2005 [DOI] [PubMed] [Google Scholar]
- 28.Hermida RC, Ayala DE: Chronotherapy with the angiotensin-converting enzyme inhibitor ramipril in essential hypertension: Improved blood pressure control with bedtime dosing. Hypertension 54: 40–46, 2009 [DOI] [PubMed] [Google Scholar]
- 29.Tobe SW, Clase CM, Gao P, McQueen M, Grosshennig A, Wang X, Teo KK, Yusuf S, Mann JF, ONTARGET and TRANSCEND Investigators : Cardiovascular and renal outcomes with telmisartan, ramipril, or both in people at high renal risk: Results from the ONTARGET and TRANSCEND studies. Circulation 123: 1098–1107, 2011 [DOI] [PubMed] [Google Scholar]
- 30.Parving HH, Brenner BM, McMurray JJ, de Zeeuw D, Haffner SM, Solomon SD, Chaturvedi N, Ghadanfar M, Weissbach N, Xiang Z, Armbrecht J, Pfeffer MA: Aliskiren Trial in Type 2 Diabetes Using Cardio-Renal Endpoints (ALTITUDE): Rationale and study design. Nephrol Dial Transplant 24: 1663–1671, 2009 [DOI] [PubMed] [Google Scholar]
- 31.Rossignol P, Cleland JG, Bhandari S, Tala S, Gustafsson F, Fay R, Lamiral Z, Dobre D, Pitt B, Zannad F: Determinants and consequences of renal function variations with aldosterone blocker therapy in heart failure patients after myocardial infarction: Insights from the Eplerenone Post-Acute Myocardial Infarction Heart Failure Efficacy and Survival Study. Circulation 125: 271–279, 2012 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.