Abstract
Context: Aldosterone production is associated with insulin resistance in obese and hypertensive subjects. However, its effect on insulin sensitivity in healthy subjects is not clear.
Objective: The objective of this study was to test the hypothesis that increased aldosterone production is associated with lower insulin sensitivity in healthy subjects.
Design: This is an analysis of data previously collected during studies conducted as part of the International Hypertensive Pathotype Consortium.
Participants and Interventions: Eighty-four subjects free of any medical or psychiatric illness were included in this study. They were studied after 7 d of a standardized high-sodium diet confirmed by 24-h urine sodium above 200 mEq. Insulin sensitivity index (ISI) was calculated after a 75-g oral glucose load with glucose and insulin measurements at 0, 30, 60, and 120 min. Serum aldosterone levels were measured after 45 min of angiotensin II (3 ng/kg/min) infusion.
Results: There were significant negative correlations between ISI and age, body mass index (BMI), diastolic blood pressure, and angiotensin II-stimulated aldosterone level (P < 0.01). On multivariate regression analysis, stimulated aldosterone level was an independent predictor of ISI after adjusting for age, BMI, and diastolic blood pressure. Stimulated aldosterone level predicted 8% of the variance in ISI (P = 0.003) with age, BMI, and diastolic blood pressure together predicting 23% of the variance in ISI. Thus, the final regression model predicted 31% of the variance in ISI (P < 0.0001).
Conclusions: Aldosterone production is associated with insulin resistance in normotensive healthy subjects independent of traditional risk factors.
Aldosterone production is associated with insulin resistance in normotensive healthy subjects independent of traditional risk factors.
Primary hyperaldosteronism is associated with insulin resistance (IR) (1). Treatment of primary hyperaldosteronism with surgery or mineralocorticoid receptor antagonists improves insulin sensitivity (1). Plasma aldosterone levels are elevated in hypertensive obese subjects (2), and one study showed an association between plasma aldosterone levels and measures of IR in hypertensive subjects (3). Aldosterone production is increased in normotensive overweight subjects and has a weak correlation with IR measured by homeostatic model assessment (HOMA-IR) (4).
The goal of the current study was to determine the relationship between aldosterone and insulin sensitivity in normotensive individuals regardless of their weight status. Because HOMA-IR is not a very sensitive measure of IR (5), we used the insulin sensitivity index (ISI) as our index of IR. The ISI derived from the oral glucose tolerance test is a better method than HOMA-IR to evaluate IR because of its better correlation with euglycemic hyperinsulinemic clamp data (6). Since aldosterone levels are influenced by multiple factors including posture, dietary sodium, and fluid intake (7), we investigated the aldosterone status under highly controlled conditions. We hypothesized that aldosterone production would be negatively associated with ISI in normotensive subjects. Furthermore, we determined the relative contribution of aldosterone to ISI compared with known predictors of IR, e.g. age, body mass index (BMI), and blood pressure.
Subjects and Methods
The data were collected during the studies conducted as part of the International HyperPath (Hypertensive Pathotype) consortium (8). Studies were conducted at three institutes: Brigham and Women’s Hospital, the University of Utah, and Vanderbilt. Data on some of these subjects have been reported before (4,8,9,10). However, there are no previous reports of an association of ISI with aldosterone levels. The institutional review board of each institution approved the studies, and all participants signed an informed written consent.
Participants were 18–65 yr of age, otherwise healthy, and not taking any drugs. Exclusion criteria included a history of diabetes mellitus, hypertension, coronary artery disease, stroke, current tobacco use, recreational drug use, alcohol intake greater than 12 ounces/wk, and any other significant medical or psychiatric illnesses. Participants were also excluded for: blood pressure above 140/90 mm Hg; abnormal serum electrolytes, thyroid, renal or liver function tests; and electrocardiographic abnormalities indicating heart block, ischemia, or prior coronary artery disease.
Participants were studied after 7 d of an isocaloric, high-sodium diet containing a minimum of 200 mmol/d sodium, 100 mmol/d potassium, 20 mmol/d calcium, and no caffeine or alcohol. Participants with urinary sodium less than 200 mmol/24 h were excluded from this study. Oral glucose tolerance test was performed in the morning in fasting subjects by administering a 75-g oral glucose load and collecting blood samples for plasma glucose and insulin at 0, 30, 60, and 120 min. After the evening meal, participants were asked to remain fasting and supine overnight for 8–10 h, and blood samples for basal aldosterone and plasma renin activity (PRA) were obtained the next morning. Angiotensin II (AngII) was infused at 3 ng/kg/min for 45 min, after which serum aldosterone was measured. Blood pressure and heart rate were monitored at baseline and every 2 min during the AngII infusion. Protocols were standardized, and all laboratory assays were performed at a central laboratory as previously described (8). Urinary and serum aldosterone levels were measured by solid phase RIA by the Coat-A -Count method (Diagnostic Products Corporation, Los Angeles, CA). The sensitivity of this method is 2.5 ng/ml (range, 2.5–120 ng/dl). The intraassay variation is 2.5–5.4%, and the interassay variation is 3.8–15.7%. AngII was measured by a solid phase double-antibody RIA (ALPCO, Windham, NH). The sensitivity of this method is 0.6 pg/ml (range, 0.6–500 pg/ml), and the precision is less than 15%. Serum cortisol was measured by Access Cortisol assay (Beckman Coulter, Chaska, MN). Urinary and serum sodium and potassium were assayed by flame photometry with an internal standard of lithium (Nova Biomedical, Waltham, MA). Urine creatinine was measured with the ACE Creatinine Reagent (Alfa Wasserman, West Caldwell, NJ). Plasma was additionally assayed for PRA, glucose, insulin, and lipids.
ISI was calculated by the method described by Matsuda and DeFronzo (6).
Statistical analyses
Histograms of the data were inspected, and normality was tested using the Shapiro-Wilk test. One subject with an ISI value of 18.7 was excluded from analysis because this value was 5 sd values from the mean and was determined to be erroneous by the laboratory. The raw data were summarized using means and sd values. BMI, serum cortisol, and AngII-stimulated aldosterone were log-transformed to achieve normality for analyses. Bivariate correlations were assessed using Pearson’s correlation coefficient. The correlations involving gender are point biserial correlations with female coded as 1 and male coded as zero. Sequential multiple variable regression models were built to assess the association between AngII-stimulated aldosterone and ISI with and without other predictors in the model. The statistical analyses were performed using SPSS for Windows version 15.0 statistical software (SPSS Inc., Chicago, IL).
Results
Eighty-four normotensive subjects were included in this study. Mean age was 40.0 ± 10.8 yr (51% men and 49% women). White race was reported by 84% of the study subjects, 9% reported being Black, and 7% Asian. Mean systolic blood pressure (SBP) was 111 ± 10 mm Hg, and diastolic blood pressure (DBP) was 68 ± 7 mm Hg. Mean BMI was 25.5 ± 3.7 kg/m2 (range, 19–36); 24 subjects (28%) had BMI above 25 kg/m2, and 13 subjects (15%) had BMI above 30 kg/m2. Other baseline characteristics were as follows: serum cortisol, 12.1 ± 3.9 μg/dl; serum sodium, 139.3 ± 4.5 mmol/liter; serum potassium, 4.1 ± 0.3 mmol/liter; serum aldosterone, 3.4 ± 2.3 ng/dl; PRA, 0.4 ± 0.4 ng/ml/h; urine sodium, 274.6 ± 63.1 mmol/24 h; and urine aldosterone, 8.2 ± 5.3 μg/24 h. ISI was 5.1 ± 2.7. Serum aldosterone levels at the end of AngII infusion were 10.5 ± 5.9 ng/dl. Serum AngII levels at the end of infusion were similar in overweight subjects with BMI above 25 kg/m2 and lean subjects with BMI below 25 kg/m2 (82.0 ± 31.4 vs. 78.4 ± 64.4 pg/ml; P = not significant).
Bivariate correlations among ISI and all potential predictors of ISI are shown in Table 1. There were significant negative correlations between ISI and age, BMI, DBP, and AngII-stimulated aldosterone (P < 0.01), but not between ISI and SBP, serum cortisol, baseline serum aldosterone, or 24-h urinary aldosterone (P > 0.15). AngII-stimulated aldosterone levels were positively associated with BMI (r = 0.31; P < 0.01) and were significantly higher in women (12.41 ± 6.48 vs. 8.63 ± 4.70 ng/dl; P < 0.01). However, women had lower SBP (108.9 ± 10.0 vs. 114.4 ± 9.7 mm Hg; P < 0.05) as well as lower DBP (65.9 ± 8.3 vs. 69.3 ± 6.3 mm Hg; P < 0.05).
Table 1.
Bivariate associations among potential predictors of ISI
| ISI | U. Aldo | Stim. S. Aldo | S. Aldo | S. Cortisol | DBP | SBP | BMI | Female | |
|---|---|---|---|---|---|---|---|---|---|
| Age | −0.38a | 0.16 | 0.04 | −0.04 | −0.10 | 0.14 | 0.09 | 0.39a | −0.15 |
| Female | −0.05 | 0.07 | 0.34b | 0.07 | −0.12 | −0.23c | −0.27c | 0.06 | |
| BMI | −0.31b | 0.22 | 0.31b | 0.16 | −0.22c | 0.23c | 0.38b | ||
| SBP | −0.15 | −0.03 | −0.06 | 0.12 | −0.02 | 0.63a | |||
| DBP | −0.30b | −0.03 | −0.06 | 0.07 | −0.12 | ||||
| S. Cortisol | 0.09 | −0.27c | −0.26c | 0.28c | |||||
| S. Aldo | −0.03 | 0.21 | 0.42a | ||||||
| Stim. S. Aldo | −0.31b | 0.45a | |||||||
| U. Aldo | −0.11 |
S. Cortisol, Baseline serum cortisol; S. Aldo, baseline serum aldosterone; Stim. S. Aldo, AngII-stimulated serum aldosterone; U. Aldo, 24-h urine aldosterone.
Correlation is significant at the 0.001 level (two-tailed).
Correlation is significant at the 0.01 level (two-tailed).
Correlation is significant at the 0.05 level (two-tailed).
Table 2 presents the results of regression models that were constructed to assess the independent contribution of AngII-stimulated aldosterone in predicting ISI after accounting for predictors that had a bivariate correlation with ISI of 0.30 or greater in absolute value (age, BMI, DBP). In every model, the additional contribution of AngII-stimulated aldosterone in predicting ISI was significant after accounting for the prior predictors. For example, in model no. 8, the reduced model containing all three established predictors of ISI accounted for 23% of the variance. When AngII-stimulated aldosterone was added to the model, the additional contribution was 8% (P = 0.003). Thus, the final model including AngII-stimulated aldosterone, age, DBP, and BMI accounted for 31% of the variance in ISI (P = 0.0001).
Table 2.
Regression analyses to assess the independent contribution of AngII-stimulated aldosterone in predicting ISI
| Model | Variable order | Variables | Partial r2 | P |
|---|---|---|---|---|
| 1 | Stim. S. Aldo | 0.10 | 0.004 | |
| 2 | 1st | BMI | 0.10 | 0.004 |
| 2nd | Stim. S. Aldo | 0.05 | 0.03 | |
| Full model | 0.15 | 0.002 | ||
| 3 | 1st | Age | 0.15 | 0.0003 |
| 2nd | Stim. S. Aldo | 0.09 | 0.003 | |
| Full model | 0.24 | 0.0001 | ||
| 4 | 1st | DBP | 0.09 | 0.005 |
| 2nd | Stim. S. Aldo | 0.11 | 0.002 | |
| Full model | 0.20 | 0.0001 | ||
| 5 | 1st | BMI and Age | 0.18 | 0.0004 |
| 2nd | Stim. S. Aldo | 0.07 | 0.01 | |
| Full model | 0.24 | 0.0001 | ||
| 6 | 1st | BMI and DBP | 0.15 | 0.001 |
| 2nd | Stim. S. Aldo | 0.07 | 0.01 | |
| Full model | 0.22 | 0.0002 | ||
| 7 | 1st | Age and DBP | 0.21 | 0.0001 |
| 2nd | Stim. S. Aldo | 0.10 | 0.001 | |
| Full model | 0.31 | 0.0001 | ||
| 8 | 1st | Age, BMI, DBP | 0.23 | 0.0001 |
| 2nd | Stim. S. Aldo | 0.08 | 0.003 | |
| Full model | 0.31 | 0.0001 |
Stim. S. Aldo, AngII-stimulated serum aldosterone.
Discussion
This study provides evidence that aldosterone production, as assessed by the response to AngII infusion, is associated with ISI in normotensive subjects when other variables are controlled. Indeed, aldosterone production is as potent a contributor to IR as the more traditional factors, i.e. BMI and DBP when assessed individually. Although AngII-stimulated aldosterone strongly correlated with baseline serum aldosterone and 24-h urinary aldosterone, these baseline assessments were not associated with IR, likely due to a lack of power for these more heterogeneous endpoints. This study was conducted under rigorously controlled conditions because aldosterone levels can be affected by multiple factors including dietary salt, posture, serum potassium, and volume status (7). The strict study protocol is likely to have contributed to the sensitivity of the study. The significance of our findings under normal physiological conditions and their clinical importance will need to be evaluated in future studies.
Although BMI was associated with increased aldosterone production, confirming our previous observation (4), it did not explain the association of AngII-stimulated aldosterone with ISI as shown by results of the multivariate regression analysis. AngII-stimulated aldosterone predicted 8% of the variability in ISI, which was roughly one third of that predicted by the other three variables together (BMI, age, and DBP). Subjects with higher BMI received higher doses of AngII due to weight-based dose calculation. However, because the measured AngII levels were similar in overweight subjects and lean subjects, it is unlikely that the higher AngII doses were responsible for higher aldosterone in insulin-resistant subjects.
Our findings are consistent with data available in patients with hypertension. Several studies have documented alterations in aldosterone production in hypertensive subjects, particularly those who have resistant hypertension (11). These alterations range from small to nondetectable increases in aldosterone on a genetic (9) or nongenetic (11) basis to definitive primary aldosteronism (10). Because IR is also common in hypertensive cohorts, several studies have assessed the relationship between aldosterone and IR and documented a positive correlation (3). Indeed, because of the particularly strong correlation between aldosterone and IR in obese hypertensive subjects, several investigators have proposed that aldosterone is the mediator of the hypertension in these patients (12,13). This hypothesis is strengthened by the reports that adipose tissue may produce aldosterone-stimulating factors that increase aldosterone production leading to hypertension (14,15). Other studies support this hypothesis by demonstrating that the high aldosterone levels in obese individuals are reduced with weight loss (13) and that in this same population, aldosterone levels correlate with insulin levels (16). Although the possibility of a chance correlation between ISI and AngII-stimulated aldosterone cannot be ruled out in our study, this scenario is unlikely because our findings are consistent with other data in this field.
More recently, it has been shown in in vitro and animal studies that aldosterone interferes with insulin signaling pathways and reduces the expression of insulin-sensitizing factors such as adiponectin and peroxisome proliferator-activated receptor-γ (17,18). Blockade of the mineralocorticoid receptor increases adiponectin and peroxisome proliferator-activated receptor-γ in adipose tissue and improves insulin sensitivity in obese, diabetic ob/ob and db/db mice (18,19). Moreover, aldosterone can induce inflammation and oxidative stress that may lead to IR by activation of serine kinases promoting phosphorylation and inactivation of insulin receptor substrate-1, thus interfering with insulin signaling pathways (20). By the same mechanisms, aldosterone may also cause pancreatic β-cell dysfunction or even apoptosis (20). Furthermore, primary hyperaldosteronism is associated with IR that improves after treatment (1). These studies raise the possibility that aldosterone can directly affect IR rather than factors associated with IR stimulating aldosterone. The present study provides support for this possibility.
In conclusion, we have demonstrated that aldosterone production is associated with IR in normotensive subjects independent of traditional risk factors. These findings add to the existing literature on the role of aldosterone as a cardiovascular risk factor.
Footnotes
This work was supported by National Institutes of Health Grants HL087060, HL47651, HL59424, and DK63214; Specialized Center of Research in Molecular Genetics of Hypertension Grant P50HL055000; and General Clinical Research Centers Grants M01RR02635, M01RR00095, and M01RR00064.
Disclosure Summary: The authors have no disclosures.
First Published Online February 3, 2010
Abbreviations: AngII, Angiotensin II; BMI, body mass index; DBP, diastolic blood pressure; HOMA-IR, homeostatic model assessment for IR; IR, insulin resistance; ISI, insulin sensitivity index; PRA, plasma renin activity; SBP, systolic blood pressure.
References
- Catena C, Lapenna R, Baroselli S, Nadalini E, Colussi G, Novello M, Favret G, Melis A, Cavarape A, Sechi LA 2006 Insulin sensitivity in patients with primary aldosteronism: a follow-up study. J Clin Endocrinol Metab 91:3457–3463 [DOI] [PubMed] [Google Scholar]
- Rossi GP, Belfiore A, Bernini G, Fabris B, Caridi G, Ferri C, Giacchetti G, Letizia C, Maccario M, Mannelli M, Palumbo G, Patalano A, Rizzoni D, Rossi E, Pessina AC, Mantero F 2008 Body mass index predicts plasma aldosterone concentrations in overweight-obese primary hypertensive patients. J Clin Endocrinol Metab 93:2566–2571 [DOI] [PubMed] [Google Scholar]
- Colussi G, Catena C, Lapenna R, Nadalini E, Chiuch A, Sechi LA 2007 Insulin resistance and hyperinsulinemia are related to plasma aldosterone levels in hypertensive patients. Diabetes Care 30:2349–2354 [DOI] [PubMed] [Google Scholar]
- Bentley-Lewis R, Adler GK, Perlstein T, Seely EW, Hopkins PN, Williams GH, Garg R 2007 Body mass index predicts aldosterone production in normotensive adults on a high-salt diet. J Clin Endocrinol Metab 92:4472–4475 [DOI] [PMC free article] [PubMed] [Google Scholar]
- García-Estévez DA, Araújo-Vilar D, Fiestras-Janeiro G, Saavedra-González A, Cabezas-Cerrato J 2003 Comparison of several insulin sensitivity indices derived from basal plasma insulin and glucose levels with minimal model indices. Horm Metab Res 35:13–17 [DOI] [PubMed] [Google Scholar]
- Matsuda M, DeFronzo RA 1999 Insulin sensitivity indices obtained from oral glucose tolerance testing: comparison with the euglycemic insulin clamp. Diabetes Care 22:1462–1470 [DOI] [PubMed] [Google Scholar]
- Dluhy RG 1978 Adrenal function testing. Compr Ther 4:53–59 [PubMed] [Google Scholar]
- Kosachunhanun N, Hunt SC, Hopkins PN, Williams RR, Jeunemaitre X, Corvol P, Ferri C, Mortensen RM, Hollenberg NK, Williams GH 2003 Genetic determinants of nonmodulating hypertension. Hypertension 42:901–908 [DOI] [PubMed] [Google Scholar]
- Pojoga L, Kolatkar NS, Williams JS, Perlstein TS, Jeunemaitre X, Brown NJ, Hopkins PN, Raby BA, Williams GH 2006 β2 Adrenergic receptor diplotype defines a subset of salt-sensitive hypertension. Hypertension 48:892–900 [DOI] [PubMed] [Google Scholar]
- Williams JS, Williams GH, Raji A, Jeunemaitre X, Brown NJ, Hopkins PN, Conlin PR 2006 Prevalence of primary hyperaldosteronism in mild to moderate hypertension without hypokalaemia. J Hum Hypertens 20:129–136 [DOI] [PubMed] [Google Scholar]
- Gonzaga CC, Calhoun DA 2008 Resistant hypertension and hyperaldosteronism. Curr Hypertens Rep 10:496–503 [DOI] [PubMed] [Google Scholar]
- Rocchini AP, Katch VL, Grekin R, Moorehead C, Anderson J 1986 Role for aldosterone in blood pressure regulation of obese adolescents. Am J Cardiol 57:613–618 [DOI] [PubMed] [Google Scholar]
- Dall'Asta C, Vedani P, Manunta P, Pizzocri P, Marchi M, Paganelli M, Folli F, Pontiroli AE 2009 Effect of weight loss through laparoscopic gastric banding on blood pressure, plasma renin activity and aldosterone levels in morbid obesity. Nutr Metab Cardiovasc Dis 19:110–114 [DOI] [PubMed] [Google Scholar]
- Ehrhart-Bornstein M, Lamounier-Zepter V, Schraven A, Langenbach J, Willenberg HS, Barthel A, Hauner H, McCann SM, Scherbaum WA, Bornstein SR 2003 Human adipocytes secrete mineralocorticoid- releasing factors. Proc Natl Acad Sci USA 100:14211–14216 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Ehrhart-Bornstein M, Arakelyan K, Krug AW, Scherbaum WA, Bornstein SR 2004 Fat cells may be the obesity-hypertension link: human adipogenic factors stimulate aldosterone secretion from adrenocortical cells. Endocr Res 30:865–870 [DOI] [PubMed] [Google Scholar]
- Goodfriend TL, Kelley DE, Goodpaster BH, Winters SJ 1999 Visceral obesity and insulin resistance are associated with plasma aldosterone levels in women. Obes Res 7:355–362 [DOI] [PubMed] [Google Scholar]
- Wada T, Ohshima S, Fujisawa E, Koya D, Tsuneki H, Sasaoka T 2009 Aldosterone inhibits insulin-induced glucose uptake by degradation of insulin receptor substrate (IRS) 1 and IRS2 via a reactive oxygen species-mediated pathway in 3T3–L1 adipocytes. Endocrinology 150:1662–1669 [DOI] [PubMed] [Google Scholar]
- Guo C, Ricchiuti V, Lian BQ, Yao TM, Coutinho P, Romero JR, Li J, Williams GH, Adler GK 2008 Mineralocorticoid receptor blockade reverses obesity-related changes in expression of adiponectin, peroxisome proliferator-activated receptor-γ, and proinflammatory adipokines. Circulation 117:2253–2261 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Hirata A, Maeda N, Hiuge A, Hibuse T, Fujita K, Okada T, Kihara S, Funahashi T, Shimomura I 2009 Blockade of mineralocorticoid receptor reverses adipocyte dysfunction and insulin resistance in obese mice. Cardiovasc Res 84:164–172 [DOI] [PubMed] [Google Scholar]
- Sowers JR, Whaley-Connell A, Epstein M 2009 Narrative review: the emerging clinical implications of the role of aldosterone in the metabolic syndrome and resistant hypertension. Ann Intern Med 150:776–783 [DOI] [PMC free article] [PubMed] [Google Scholar]