Changes in serum aldosterone are associated with changes in obesity-related factors in normotensive overweight and obese young adults

Abstract

Recent data suggest excess circulating aldosterone promotes cardiometabolic decline. Weight loss may lower aldosterone levels, but little longitudinal data is available in normotensive adults. We aimed to determine if, independent of changes in sodium excretion, reductions in serum aldosterone are associated with favorable changes in obesity-related factors in normotensive overweight/obese young adults. We studied 285 overweight/obese young adult participants (body mass index (BMI) ≥ 25 and < 40 kg/m², age 20–45 years) in a clinical trial examining the effects of a one year diet and physical activity intervention with or without sodium restriction on vascular health. Body weight, serum aldosterone, 24-hr sodium and potassium excretion, and obesity-related factors were measured at baseline, 6, 12, and 24 months. Weight loss was significant at 6 (7%), 12 (6%), and 24 months (4%) (all p<0.0001). Decreases in aldosterone were associated with decreases in C-reactive protein, leptin, insulin, homeostasis assessment of insulin resistance, heart rate, tonic cardiac sympathovagal balance, and increases in adiponectin (all p<0.05) in models adjusting for baseline age, sex, race, intervention arm, time since baseline, and sodium and potassium excretion. Weight loss and reductions in thigh intermuscular fat (IMAT) were associated with decreases in aldosterone in the subgroup (n=98) with metabolic syndrome (MetS) at baseline (MetS x weight loss p=0.04, MetS x change in IMAT p=0.04). Favorable changes in obesity-related factors are associated with reductions in aldosterone in young adults with no risk factors besides excess weight, an important finding given aldosterone’s emergence as an important cardiometabolic risk factor.

INTRODUCTION

Aldosterone plays important roles in blood pressure regulation and sodium and water balance, but inappropriately elevated levels have been found to promote cardiovascular decline ¹ and are also associated with obesity-related metabolic abnormalities^2–6. Furthermore, several studies have found that aldosterone levels are higher in overweight/obese individuals, particularly those with excess visceral fat^2–5. Aldosterone was found to decrease with modest weight loss when individuals reduced their calorie intake while maintaining a low, moderate or high dietary sodium intake ^{5, 7–10}.

These findings may be partially explained by adipose tissue production of renin-angiotensin-aldosterone system (RAAS) components ^11–13. Human adipocytes also produce mineralocorticoid-stimulating factors that increase adrenal aldosterone secretion independently of angiotensin II (AngII) or potassium ¹⁴. Another mechanism for RAAS over-activition in overweight/obese individuals is impaired renal sympathovagal balance, which stimulates renin release by the kidneys ¹¹. Furthermore, both renal and cardiac sympathovagal balance are worsened by excess circulating aldosterone ¹⁵. Finally, increased formation of Ang II by large insulin-resistant adipocytes inhibits the recruitment and differentiation of preadipocytes, which leads to ectopic fat storage and decreased insulin sensitivity ¹⁶. Altogether, this evidence suggests the presence of a vicious cycle wherein excess adiposity promotes aldosterone production and excess aldosterone, along with other RAAS components, drives adipose inflammation, insulin resistance, and cardiovascular decline.

Though several studies have reported decreases in aldosterone with weight loss ^{5, 7–10}, no study of healthy normotensive young adults has examined associations between changes in aldosterone and changes in obesity-related factors while accounting for changes in sodium intake, an important determinant of circulating aldosterone levels. We hypothesized that, independent of changes in 24-hour sodium excretion, reductions in serum aldosterone would be associated with weight loss and reductions in regional adiposity, thigh intermuscular adiposity, inflammation, leptin, insulin resistance, and sympathovagal balance, and increases in adiponectin and ghrelin in normotensive overweight/obese young adults followed over the course of a one year lifestyle intervention and one year post-intervention period. We also hypothesized that, independent of weight loss, decreases in both serum aldosterone and sodium excretion would be associated with decreases in blood pressure. Finally, because individuals with greater metabolic dysfunction may be more sensitive to the cardiovascular and metabolic effects of excess aldosterone and dietary sodium ¹⁷, we hypothesized that individuals with metabolic syndrome (MetS) would show stronger associations between the factors of interest.

METHODS

Study Population

Subjects came from the Slow Adverse Vascular Effects of excess weight study (SAVE), a randomized-controlled trial (NCT00366990) evaluating the effects of weight loss, increased physical activity, and reduced dietary sodium intake on vascular health. Participants were recruited during June 2007 - May 2009 using mass mailing. The study was approved by the University of Pittsburgh IRB and all participants provided written informed consent.

Eligible participants were men and women 20–45 years of age who were overweight/obese (body mass index (BMI) 25–39.9 kg/m²) and physically inactive (<8 months of physical activity during the past 12 months). Exclusions included 1) diabetes, 2) hypertension or average screening blood pressure ≥140/90 mmHg, 3) cholesterol lowering, anti-psychotic, or vasoactive medication use and 4) current pregnancy or lactation. Three hundred and forty-nine eligible participants received a one-year lifestyle intervention promoting diet and PA. Participants were randomized to either 1) diet and physical activity alone (Control Na/lifestyle) or to 2) diet and physical activity plus reduced sodium intake (Low Na/lifestyle). The lifestyle intervention was delivered in group sessions that occurred weekly for months 1–4, biweekly for months 5–8, and monthly for months 9–12. The goal of the intervention was 10% weight loss over 6 months and continued weight maintenance thereafter. The goal of the sodium reduction intervention (Low Na) was to gradually reduce sodium intake to approximately 1 mg Na/kcal/day. Participants were to be seen at screening, baseline, and 6, 12, and 24 months following randomization.

Demographic and Physical Measures

Age, race, and smoking status were self-reported. Race was re-coded as black vs. non-black. Smoking status was assessed as current vs. past or never. Weight was measured in kilograms using a balance scale. Height was measured in centimeters using a stadiometer. Waist circumference was measured against the participant’s skin at the narrowest part of the torso between the ribs and the iliac crest. Blood Pressure (BP) was measured with a mercury sphygmomanometer after participants sat quietly for 5 minutes with feet flat on the floor. Final BP was the average of the last 2 of 3 readings taken 30 seconds apart.

Blood and Urine Assays

Blood analytes were measured at the Heinz Laboratory at the University of Pittsburgh’s Graduate School of Public Health using standard methods as previously described ^{18, 19}. Briefly, blood specimens were obtained between 0700 and 1130 h from upright subjects after a fasting period of at least 9 hours. Serum aldosterone was measured using an enzyme-linked immunoassay developed by Diagnostic Systems Laboratories, Inc. (Webster, TX, USA). The intra- and inter-assay CV% for insulin were 4.8% and 10.5% respectively. The CV% for the other assays were all <3%.

Twenty-four hour urine collections were performed within 2 weeks of the clinic visits at which all other measurements were determined. Valid collections had volume between 500 and 4000 mL, duration ≥22 and ≤26 hours, and creatinine within the expected range ²⁰. Sodium, potassium, and creatinine were determined as previously described ¹⁸.

Regional Measures of Adiposity

At baseline and 12 months, single-slice computed tomography (CT) scans of the abdomen and thigh were acquired using a C-150 Ultrafast CT Scanner (GE Imatron, San Francisco, CA). Slice thickness was 6 mm. Abdominal scans were transverse images between L4 and L5 obtained during suspended respiration; left thigh images were transverse images 15 cm above the patellar apex.

CT images were interpreted by one reader (MBC) using Slice-O-Matic software. A pixel range of −30 to −190 Hounsfield units was used to define fat in the scan circumference. Areas were calculated by multiplying the number of pixels of a given tissue type by the pixel area. Density values were determined by averaging the CT number (pixel density) values of the regions outlined on the images. For the abdominal scan, region of interest lines were drawn along fascial planes. Fat above the internal fascial plane was considered subcutaneous fat and fat below the plane was considered visceral fat area. For the thigh scan, a single region of interest line was drawn along the deep fascial plane surrounding the thigh muscle. Fat above this line was considered subcutaneous fat, and fat below was considered intermuscular fat ²¹.

Heart Rate Variability

At baseline and 6 months, an ANSAR monitor (ANX-3.0, ANSAR Group Inc., Philadelphia, PA) provided continuous and noninvasive measurements of electrocardiogram signals (for heart rate variability (HRV) assessment) and bioimpedance plethysmography signals (for respiratory rate variability (RRV) assessment) ²². A spectral analysis of the HRV and RRV was generated using ANSAR software. The low-frequency area (LFa) was centered on the HRV spectrum from 0.04 – 0.10 Hz, which is taken to reflect sympathetic cardiac activity. From the spectral analysis of the RRV, the frequency of the peak mode was defined as the fundamental respiratory frequency (FRF). A 0.12 Hz wide window from the HRV spectrum was centered at the FRF and used to generate the respiratory frequency area (RFa), which is taken to reflect parasympathetic cardiac activity ²³. The area under the spectral curve centered on the FRF is computed as RFa. The remaining area under the spectral curve in the low-frequency bandwidth is computed as LFa. The measure examined in this analysis was LFa/RFa during an initial 5 minute resting period (tonic sympathovagal balance).

Statistical Methods

Whether changes in variables of interest were statistically significant was determined by testing the coefficient for time in linear mixed models with unstructured error covariance. Intervention arm was included as a covariate for consistency with trial design. An interaction between intervention arm and time was used to test whether changes differed by intervention arm.

The main analysis began with linear mixed models with aldosterone as the time-varying dependent variable, measured at baseline and 6, 12, and 24 months. Independent variables were age, sex, race, baseline and within-subject changes in sodium and potassium excretion, and baseline and within-subject changes in obesity-related variables of interest. The following obesity-related variables were individually evaluated: BMI, weight, waist circumference, abdominal visceral and subcutaneous adipose tissue areas, thigh intermuscular adipose tissue area (IMAT), insulin, HOMA-IR, adiponectin, leptin, ghrelin, CRP, resting supine heart rate, and sitting LFa/RFa (cardiac sympathovagal balance). A fixed quadratic time effect and random intercepts and linear and quadratic time effects were evaluated and included if found to be significant at p<0.10. To determine whether associations varied over time or across subgroups of interest, interactions between changes in the obesity-related factors and time, race, sex, age, or the presence of MetS ²⁴ at baseline were tested. To evaluate whether associations between aldosterone and non-anthropometric obesity-related factors were independent of weight loss and changes in insulin levels, baseline and within-subject changes in BMI, weight, waist circumference, or fasting insulin were added to the models.

Linear mixed models for systolic (SBP) and diastolic (DBP) blood pressure were used to assess whether changes in urinary sodium excretion or serum aldosterone were associated with BP changes independent of weight loss, adjusting for baseline age, sex, race, weight, and baseline and within-subject changes in potassium excretion. Interactions between changes in sodium excretion and baseline serum aldosterone, race, sex, age, and MetS status were evaluated. Finally, both in the models for aldosterone and the models for BP, each component of the MetS ²⁵ was investigated individually in place of overall MetS status, using a Bonferroni correction for multiple comparisons.

Sensitivity analyses were performed to evaluate potential effects of the missing data. First, sodium/creatinine and potassium/creatinine excretion ratios were used in place of 24-hr sodium and potassium excretion in order to include data from all urine collections rather than only completely valid collections. Second, to examine the hypothesis that participants with missing follow-up data were on average less successful in achieving weight loss than participants with complete data, pattern-mixture modeling and multiple imputation were used. Multiple imputation was performed for each missing data pattern under the assumption that the multivariate distribution of the missing data for each pattern, given the observed data, followed the corresponding distribution in those subjects with complete data who had achieved less than the mean weight loss at follow-up visits with unavailable data for that pattern. P values ≤0.05 were considered statistically significant. Statistical analyses were performed using SAS (Statistical Analysis Software v9.3, Cary, NC).

RESULTS

The study population consisted of 285 participants in the SAVE trial who provided valid baseline 24-hr urine collections and serum aldosterone data. The sample had a mean age of 38.4 years (SD 5.8) at baseline. Twenty percent of participants were male and 15% black. Additionally, 8% self-identified as current smokers. Mean values of key clinical characteristics over the course of the intervention are shown in Table 1. Average weight loss was 7.1% at 6 months, 6.4% at 12 months, and 3.5% at 24 months.

Table 1.

Clinical Characteristics across the One Year Intervention and at One Year Post-Intervention

Characteristic	Baseline (N=285)	6 Months (N=233)	12 Months (N=210)	24 Months (N=189)
Aldosterone (pg/mL)	108 (79, 156)	117 (84.3, 156)	104 (84, 140)	107 (83.1, 157.5)
Weight (kg)	91.8 (13.3)	84.6 (13.3)*	85.1 (14.2)*	88.1 (14.6)*
BMI (kg/m2)	32.9 (3.7)	30.3 (4.1)*	30.4 (4.4)*	31.4 (4.5)*
Waist Circumference (cm)	100.1 (10.3)	94.8 (10.7)*	95.0 (11.8)*	97.5 (12.3)*
SBP (mmHg)	113.2 (10.1)	109.5 (9.0)*	109.7 (9.6)*	112.2 (9.9)
DBP (mmHg)	72.7 (8.5)	70.6 (8.2)*	71.7 (7.8)	73.8 (9.1)*
Glucose (mg/dL)	97.3 (7.9)	97.8 (8.4)	98.2 (8.3)	97.6 (9.5)
Insulin (μU/mL)	12.5 (9.4, 16.7)	11.4 (8.7, 15.4)*	11.8 (9.3, 15.3)	11.9 (9.5, 15.4)
HOMA-IR (mmol/L x μU/mL)	3.0 (2.2, 4.2)	2.7 (2.1, 3.8)*	2.9 (2.2, 4.0)	2.9 (2.2, 3.8)
LDL-C (mg/dL)	123.2 (32.8)	121.4 (29.1)	123.3 (30.8)	125.5 (32.6)
HDL-C (mg/dL)	52.5 (12.7)	52.8 (12.1)	55.5 (13.7)*	54.5 (13.3)*
Triglycerides (mg/dL)	115.5 (79, 169.5)	94 (67, 135)*	88 (71, 136)*	99 (75, 146)*
CRP (mg/L)	2.6 (1.4, 5.6)	2.2 (1.0, 4.6)*	2.1 (0.94, 4.2)*	2.3 (0.91, 5.0)*
Leptin (ng/mL)	26.2 (13.1)	18.6 (11.5)*	21.0 (13.5)*	22.7 (13.4)*
Adiponectin (μg/mL)	11.9 (6.1)	12.1 (5.6)*	12.1 (5.7)	10.6 (5.7)*
Ghrelin (pg/mL)	673.5 (547, 874.5)	774 (614, 1042)*	804.5 (629, 1121.5)*	875 (711, 1113)*
Sodium Excretion (mmol/24hr) #	185.8 (69.1)	154.5 (65.2)*	156.9 (58.9)*	157.6 (63.5)*
Potassium Excretion (mmol/24hr) #	60.7 (22.1)	62.0 (21.0)	63.9 (23.1)	61.6 (21.3)
Heart Rate (beats/min)	64.3 (9.2)	62.7 (8.4)*	64.0 (8.9)	63.6 (8.8)
Sitting cardiac sympathovagal balance^	1.6 (0.87, 2.9)	1.1 (0.64, 2.6)*	-------	-------
Abdominal visceral fat area (cm2)$	117.8 (56.0)	-------	99.1 (53.5)*	-------
Abdominal subcutaneous fat area (cm2)$	425.0 (122.4)	-------	361.0 (132.2)*	-------
Thigh intermuscular fat area (cm2)$	13.0 (4.8)	-------	7.7 (3.7)*	-------

Mean (SD) or median (IQR) are shown.

^*P<0.05 versus baseline in a linear mixed model with time since baseline as a nominal variable and with adjustment for intervention arm. Aldosterone, insulin, HOMA-IR, triglycerides, CRP, ghrelin, and sitting LFa/RFa were log transformed for modeling.

SBP=systolic blood pressure, DBP=diastolic blood pressure, LDL-C=low density lipoprotein cholesterol, HDL-C=high density lipoprotein cholesterol, HOMA-IR= homeostasis model assessment of insulin resistance, CRP=C-reactive protein, LFa=low frequency area, RFa=respiratory frequency area.

^#Baseline N=285, 6 Months N=184, 12 Months N=158, 24 Months N=136.

^{^}Baseline N=277, 6 Months N=227.

^$Baseline N=272, 12 Months N=200. To convert to SI units, use the following conversion factors: glucose, 0.0555; insulin, 6.945; LDL-C and HDL-C, 0.0259; triglycerides, 0.0113; CRP, 9.524; leptin, 0.0625; ghrelin, 0.296; aldosterone, 2.774

The only measures that differed at least marginally by intervention arm were changes in sodium excretion and serum aldosterone. Mean sodium excretion was decreased from baseline by 48.1 mmol/24hr (SD 79.7) at 6 months, 35.0 mmol/24hr (SD 80.5) at 12 months, and 42.0 mmol/24hr (SD 75.8) at 24 months in the Low Na/lifestyle arm, but decreased by only 9.1 mmol/24hr (SD 77.1) at 6 months, 20.5 mmol/24hr (SD 84.8) at 12 months, and 7.6 mmol/24hr (SD 78.2) at 24 months in the Control Na/lifestyle arm (p<0.001, p=0.27, and p=0.01 respectively for between-arm comparisons). Serum aldosterone was marginally higher in the Low Na/lifestyle arm compared to the Control Na/lifestyle arm at 6 months only (p=0.06).

Changes in weight, BMI, waist circumference, and abdominal adiposity were not associated with changes in aldosterone in multivariable mixed models for log aldosterone, though there was a marginal association between decreased thigh IMAT and decreased aldosterone (Table 2). Changes in circulating adipokines and markers of insulin resistance, inflammation, and tonic cardiac sympathovagal balance were strongly associated with changes in aldosterone. As expected, there were at least marginal associations between increases in aldosterone and both decreases in sodium excretion and increases in potassium excretion in all models (p<0.10 for all, data not shown).

Table 2.

Associations between Changes in Serum Aldosterone and Changes in Obesity-Related Factors Over the Course of the Study

Independent Variable (Change from Baseline)	Percent change in aldosterone associated with one unit change in independent variable (95% CI)	P1
Anthropometric measures
Weight (%)	0.029 (−0.50, 0.56)	0.92
BMI (kg/m2)	−0.007 (−1.60, 1.61)	0.99
Waist Circumference (cm)	0.12 (−0.45, 0.69)	0.69
Serum measures
Insulin (μU/mL)	1.41 (0.64, 2.19)	0.0005
CRP (mg/L)	2.33 (1.11, 3.56)	0.0002
Leptin (ng/mL)	0.62 (0.21, 1.04)	0.003
Adiponectin (μg/mL)	−1.98 (−3.01, −0.94)	0.0002
Ghrelin (pg/mL)	0.004 (−0.01, 0.01)	0.42
Cardiac measures
Heart Rate (beats/min)	0.91 (0.36, 1.47)	0.001
Log sitting cardiac sympathovagal balance	11.40 (5.04, 18.15)	0.0004
CT adiposity measures
Abdominal visceral fat area (cm2)	0.12 (−0.10, 0.34)	0.28
Abdominal subcutaneous fat area (cm2)	0.04 (−0.04, 0.12)	0.29
Thigh intermuscular fat area (cm2)	2.22 (−0.15, 4.66)	0.053

¹P values are from linear mixed models for log aldosterone (pg/mL) that included baseline age, sex, race (black/non-black), intervention arm, time since baseline, and baseline and within-subject changes in 24-hr urinary sodium and potassium excretion, and baseline and within-subject changes in the specified independent variable. CRP=C-reactive protein. The association with HOMA-IR (homeostasis model assessment of insulin resistance) (p=0.02) was similar to the association with fasting insulin and thus is not shown. Percentage changes in aldosterone for a one unit change in the independent variable were calculated using the formula 100*(exp(β) −1), where β was the parameter estimate for the independent variable from the mixed model. To convert to SI units, use the following conversion factors: insulin, 6.945; CRP, 9.524; leptin, 0.0625; ghrelin, 0.296.

Weight loss (Figure 1a) and BMI reduction were associated with reduced aldosterone in the subgroup (n=98, 34%) with MetS at baseline (p<0.05 for both interactions). However, interactions between MetS and changes in waist circumference or abdominal adipose tissue depots were not significant (p>0.10 for all), though decreased thigh IMAT was associated with decreased aldosterone in the subgroup with MetS (Figure 1b). When each MetS component was investigated individually in the models for aldosterone, no component showed a significant interaction with changes in obesity-related factors or sodium or potassium excretion. The associations between changes in aldosterone and changes in all non-anthropometric obesity-related factors were unaltered by additional adjustment for baseline and within-subject changes in weight, BMI, waist circumference, fasting insulin, or HOMA-IR (data not shown).

Figure 1.

Neither changes in sodium excretion nor changes in aldosterone were associated with changes in BP (Table 3). However, the association between reduced sodium excretion and reduced DBP was marginally greater in subjects who had MetS at baseline or who were of black race (Figures 2 and 3). These interactions were not significant for SBP (p>0.20 for both). Again, no significant interactions were detected between individual MetS components and changes in sodium excretion (data not shown).

Table 3.

Associations between Changes in Blood Pressure and Changes in Weight, Serum Aldosterone, and Urinary Electrolytes Over the Course of the Study

Independent Variable	Parameter Estimate	Standard Error	P1
Diastolic Blood Pressure
Weight loss (%)	0.20	0.045	<0.0001
Change in sodium excretion (mmol/24hr)	0.0075	0.0047	0.11
Change in potassium excretion (mmol/24hr)	−0.0059	0.016	0.72
Change in log aldosterone (pg/mL)	−0.35	0.67	0.60
Systolic Blood Pressure
Weight Loss (%)	0.28	0.047	<0.0001
Change in sodium excretion (mmol/24hr)	0.0031	0.0049	0.52
Change in potassium excretion (mmol/24hr)	−0.013	0.017	0.45
Change in log aldosterone (pg/mL)	−0.85	0.70	0.23

¹P values are from linear mixed models for each respective blood pressure measure that included baseline age, sex, race (black/non-black), intervention arm, time since baseline, baseline weight, percent weight loss, and baseline and within-subject changes in serum aldosterone and 24-hr urinary sodium and potassium excretion.

Figure 2.

Figure 3.

In sensitivity analyses, when sodium/creatinine and potassium/creatinine excretion ratios were used in place of 24-hr sodium and potassium excretion, 55 additional subjects were included. However, the associations of interest were similar to those from the original models (data not shown). Results from pattern-mixture modeling, with missing follow-up data multiply imputed, also differed little from those in the original mixed models, though most associations were slightly weaker under the assumption of less successful weight loss among dropouts than completers (Supplemental Data available online).

DISCUSSION

The main findings of this study were that, independent of changes in sodium excretion, decreases in serum aldosterone were associated with reductions in fasting insulin, HOMA-IR, CRP, leptin, heart rate, and cardiac sympathovagal balance and increases in adiponectin in normotensive overweight/obese young adults during a one year lifestyle intervention and one year post-intervention period. In addition, though changes in weight and aldosterone were unassociated in the total sample, such an association was evident in the subgroup with MetS at baseline. Finally, decreased intermuscular thigh fat was marginally associated with decreased aldosterone in the total sample and significantly associated with decreased aldosterone in the subgroup with MetS at baseline. These findings are important because this is the first study in a large sample of overweight/obese otherwise healthy young adults to report associations between changes in circulating aldosterone and changes in a wide variety of obesity-related factors during and following a behavioral weight loss intervention.

Previous small or moderate sized studies have reported decreases in serum aldosterone or aldosterone excretion with modest weight loss in obese postmenopausal women ⁸, young overweight/obese adults ^{5, 7}, middle-aged overweight/obese adults ^{9, 26}, and obese adults who submitted to very low calorie diets ^{5, 10}. Some studies have reported significant associations between reductions in circulating RAAS components and reductions in central adiposity ^{5, 8} or insulin resistance ⁵ during weight loss, though not all studies agree ⁷. Potential reasons for discrepant findings between studies include different levels of sodium intake and heterogeneous study populations. Unlike the present study, no past studies consisted of only normotensives, evaluated the effect of weight loss on aldosterone independent of changes in discretionary sodium intake, followed participants after weight loss, or examined as many obesity-related factors.

The present findings are particularly important given the emergence of aldosterone as a cardiometabolic risk factor, promoting not only hypertension, but also inflammation and remodeling of the heart, vasculature, kidneys, and adipose tissue ²⁷. Higher levels of circulating aldosterone predict incident hypertension ²⁸ and metabolic syndrome ⁶ in the general population. Furthermore, this study and several others have found that higher aldosterone levels correlate with greater insulin resistance ^{3, 5}, an association that may be independent of anthropometric measures of body size ³. Thus, it appears that aldosterone may influence cardiometabolic health independently of BMI and other traditional risk factors.

A likely explanation for the stronger associations of circulating aldosterone with markers of metabolic dysfunction than with anthropometric measures of body size or abdominal adipose tissue area is that it is the quality rather than the quantity of adipose tissue that determines cardiometabolic dysfunction ^{11, 29}. ‘Dysfunctional’ adipose tissue is characterized by hypertrophied adipocytes, increased macrophage infiltration, hypoxia, and marked changes in adipokine and free fatty acid secretion ^{29, 30}. Elevated production of leptin, angiotensinogen, and reduced production of adiponectin accompany the accumulation of dysfunctional fat ²⁹. In addition, as excess energy-intake overwhelms the body’s fat storage capacity, ectopic fat is stored in skeletal muscle and the liver ²⁹. These changes promote insulin resistance, chronic systemic inflammation, RAAS activation, sympathoactivation, and oxidative stress ²⁹. In obese individuals with metabolic dysfunction, both visceral and subcutaneous adipose tissue depots are characterized by increased proinflammatory macrophage content and adipocyte hypertrophy ³¹. These morphological changes are not necessarily accompanied by significant changes in the amount of total body fat or visceral or subcutaneous fat mass ³¹, but they are linked to intrahepatic and intramuscular fat storage, which promote metabolic abnormalities ²¹. To our knowledge, this is the first study to report associations between serum aldosterone and intermuscular fat.

Although it is impossible in this study to determine which obesity-related factors most influenced serum aldosterone or which factors were most influenced by serum aldosterone, it is likely that all of the investigated factors are both causes and consequences of cardiometabolic decline ³². Adipocytes produce angiotensinogen and Ang II, contributing to elevated circulating levels of these hormones in obese individuals ^{33, 34}. Though there is argument that human adipocytes do not produce aldosterone ¹², a recent study suggests otherwise ¹³. Additionally, it was recently discovered that several adipocyte-derived factors increase adrenal aldosterone production independent of Ang II and serum potassium ^{14, 35, 36}. Type 1 and 2 adiponectin receptors are also present in the adrenal cortex, and may influence aldosterone secretion ³⁷. Furthermore, increased aldosterone secretion by adrenocortical cells results in greater binding and activation of adipocyte mineralocorticoid receptors, which in turn impacts adipose differentiation, expansion, and inflammation ³². Finally, there is evidence that both circulating and adipose RAAS are influenced by autonomic activity. Sympathetic nerve stimulation increases the release of renin and Ang II and the stimulatory effect of Ang II on adrenal aldosterone secretion ³⁸. Elevated circulating aldosterone also induces cardiac and renal sympathetic activation ¹⁵. Though associations between reductions in weight and blood pressure were evident in this study, changes in aldosterone and sodium were not associated with changes in blood pressure. It may be that chronically elevated aldosterone increases blood pressure over longer time periods, such as the four years over which persons were followed in a study that found aldosterone to predict incident hypertension ²⁸. In addition, the effect of weight loss may have overwhelmed the effects of concurrent changes in sodium and aldosterone. However, the associations between reductions in sodium excretion and reductions in blood pressure in individuals of black race or who had MetS agree with past studies ^{39, 40} and suggest that these subgroups may particularly benefit from sodium reduction along with weight loss to reduce blood pressure.

Limitations and Strengths

There were several limitations to this study. Because the focus of this study was not a comparison of the randomized treatment arms, it was not possible to determine causal relationships. However, the longitudinal design of this study did minimize the influence of time-independent confounders. Another limitation was the <10% mean weight loss achieved by study subjects. This smaller than expected weight loss may have limited our ability to detect associations of interest. Small numbers of males and non-whites provided insufficient power to stratify these analyses by sex or race. In addition, there is some evidence that aldosterone levels may differ by sex and race ⁴¹; thus the use of a heterogeneous sample that included 20% male and 15% black subjects may have diluted our findings. In addition, aldosterone levels vary throughout the menstrual cycle. Importantly however, we found no statistically significant interactions with sex or race in the models for serum aldosterone, suggesting that the associations between aldosterone and obesity-related factors did not differ by sex or race. Another limitation was that the lack of data on other RAAS components and natriuretic peptides prevented us from examining the extent to which changes in these factors contributed to the detected associations. However, the evidence that excess adiposity directly stimulates mineralocorticoid secretion ^{14, 35, 36} suggests that renin and other established drivers of aldosterone secretion may not fully explain the associations detected in this study. Furthermore, hypertension and the use of anti-hypertensives, the most common stimuli for high levels or changes in renin release, were not present in any study subjects, and in all analyses, we adjusted for baseline levels and changes in sodium excretion, a major stimulus of the RAAS. It could be that a measure of 24-hour aldosterone excretion might have reflected chronic circulating aldosterone exposure more accurately than a serum measurement. Another limitation was the missing data; however sensitivity analyses suggested that the findings were robust and likely not biased by the missing data. A strength of this study was that no participants were using antihypertensive or vasoactive medications, thus eliminating treatment related confounding. Finally, the large variety of measured obesity-related factors and the longitudinal design of this study provided novel insights into the complex role of aldosterone in cardiometabolic health during and after lifestyle modification.

Conclusions

In conclusion, in normotensive, overweight/obese young adults followed over the course of a one year lifestyle intervention and one year follow-up period, reductions in fasting insulin, HOMA-IR, CRP, leptin, heart rate, tonic cardiac sympathovagal balance, and increases in adiponectin are associated with decreases in serum aldosterone. Additionally, 5% weight loss is associated with a 4% reduction in serum aldosterone and a reduction in intermuscular fat of 5 cm² is associated with an 18% reduction in serum aldosterone in individuals with MetS. These findings, along with recent studies showing that mineralocorticoid receptor antagonists improve adipose tissue function in animal models of obesity ⁴², suggest that future trials should test the efficacy of these drugs for reducing cardiometabolic risk in overweight/obese individuals, particularly in those with metabolic abnormalities. Of course, positive lifestyle changes must continue to be recommended to all persons with excess weight, as even modest weight loss can improve cardiometabolic health.