ENAC blockade reduces blood pressure and arterial stiffness in adults with obesity and insulin resistance

Abstract

Obesity and insulin resistance promote arterial stiffening and hypertension, increasing cardiovascular risk. Activation of the epithelial sodium channel (ENAC) contributes to vascular stiffening in preclinical models, but the vascular effects of ENAC inhibition in adults with obesity and insulin resistance are not well defined. In this Phase II, 24-week, randomized, double-blind, single-center, placebo-controlled trial, 137 adults aged 30-70 years with overweight or obesity and at least one metabolic syndrome feature were randomized (2:1) to the ENAC inhibitor amiloride (5 mg daily) or placebo. Carotid-femoral pulse wave velocity (cfPWV), blood pressure, and vascular function were assessed at baseline, 12 weeks, and 24 weeks. Amiloride significantly reduced arterial stiffness, with decreases in cfPWV at 12 and 24 weeks, whereas no changes were observed with placebo. Systolic blood pressure was also reduced, with a mean reduction of 5.6 mmHg at 24 weeks. Older age was associated with greater reductions in cfPWV and systolic blood pressure. Amiloride increased serum potassium and lowered fasting glucose but did not significantly affect brachial artery flow-mediated dilation. No severe adverse events were observed. In conclusion, low-dose amiloride improves blood pressure and arterial stiffness in adults with overweight or obesity and features of metabolic syndrome, without major safety concerns. These findings suggest that blood pressure lowering with amiloride is associated with favorable changes in vascular stiffness in this population.

Graphical Abstract

NEW & NOTEWORTHY

Obesity and insulin resistance accelerate arterial stiffening and hypertension, increasing cardiovascular risk. Activation of the epithelial sodium channel (ENAC) contributes to vascular stiffening in preclinical models, yet the vascular effects of ENAC inhibition in adults with obesity and insulin resistance remain poorly characterized. Here, we demonstrate that low-dose amiloride reduces blood pressure and improves arterial stiffness in adults with overweight or obesity and features of metabolic syndrome, without major safety concerns.

INTRODUCTION

Obesity leads to the development of cardiovascular disease by inducing vascular abnormalities such as arterial stiffening, hypertension, and endothelial dysfunction.¹ One well-characterized mechanism underlying the heightened cardiovascular risk associated with obesity and insulin resistance is activation of the mineralocorticoid receptor (MR).² Activation of MR signaling initiates distinct remodeling and inflammatory pathways in both the vasculature and renal tissue, processes closely linked to the development of hypertension and increased arterial stiffness.² MR activation increases the expression and activity of the epithelial sodium channel (ENAC) in renal tubules and the vasculature, although other hormones, such as vasopressin and insulin, can also modulate its activity.^{3, 4}

In preclinical models, ENAC activation increases vascular stiffness, reduces vasodilation, and promotes inflammation, contributing to arterial stiffening.⁵ Mouse studies further show that obesogenic conditions enhance vascular ENAC activation, especially in females. ^6–8 These findings suggest that pharmacological inhibition of ENAC may serve as a therapeutic approach to reduce obesity-related cardiovascular risk. Amiloride, a potassium-sparing diuretic and ENAC inhibitor, is currently used as an adjunctive treatment for hypertension and heart failure. Still, its potential to improve vascular function in obesity and metabolic syndrome has not been systematically studied.⁹

Accordingly, the objective of this phase II, randomized, double-blind, single-center, placebo-controlled clinical trial was to evaluate the effects of amiloride on vascular function in adults with overweight or obesity and features of metabolic syndrome. The primary endpoint was the change in carotid-femoral pulse wave velocity (cfPWV), a gold-standard measure of central arterial stiffness, over a 24-week treatment period. Secondary endpoints included changes in systolic and diastolic blood pressure (SBP and DBP), brachial artery flow-mediated dilation (FMD), as an indicator of endothelial function, and metabolic parameters.

METHODS

Ethics Approval and Participants

All human study procedures conformed to the Declaration of Helsinki and were approved by the University of Missouri Institutional Review Board (protocol 2012990) and an independent Data Safety Monitoring Committee. The double-blinded, randomized, single-center, placebo-controlled clinical trial was registered at ClinicalTrials.gov (NCT03837626). Written informed consent was obtained from all subjects before any study procedures were completed. Women and men aged 30 to 70 years with overweight or obesity, defined as a body mass index (BMI) of 25.1 to 50 kg/m² or a waist circumference >88 cm (>35 in) for women and >102 cm (>40 in) for men, fasting insulin >10 mU/L, and at least one additional characteristic of metabolic syndrome, including elevated triglycerides (≥150 mg/dL); reduced HDL cholesterol (<40 mg/dL in men or <50 mg/dL in women); elevated blood pressure (≥130/85 mmHg or current antihypertensive treatment); or elevated fasting glucose (≥100 mg/dL) were included. Exclusion criteria included history of diabetes, known cardiovascular events within the last 12 months, history of uncontrolled thyroid disease, cirrhosis or estimated glomerular filtration rate <50 mL/min, use of potassium sparing medication or chronic NSAIDs, active cancer, excessive alcohol consumption (>7 weekly servings per week for women and >14 weekly servings per week for men), current tobacco use, non-controlled hypertension, participation in regular exercise, serum potassium >5mEq/L or blood pressure <110/70 mmHg at the time of screening.

Study Procedures

Individuals were randomly assigned in a 2:1 ratio to receive either amiloride 5 mg or a placebo, using a Web-based system (Research Randomizer)¹⁰ with stratification by age (younger than and older than 51 years) in males and menopausal status in females. The University of Missouri Hospital Investigational Pharmacy prepared and administered the amiloride/placebo capsules. Dose reduction to 2.5 mg daily was permitted per protocol in cases of potassium >5.5 mEq/L; one individual required such a dose reduction.

On the experimental visit days, research subjects were admitted to the Clinical and Translational Science Unit at the University of Missouri in the morning after an overnight fast and refraining from morning medications, caffeine, alcohol, multivitamins, and vigorous physical activity for 12 hours. Scheduling of experimental visits in females was not controlled for the menstrual cycle, given the duration of the intervention. Participants were also instructed to maintain their usual diet during the intervention. Upon arrival, subjects underwent anthropometric measurements including height, weight, and body composition via dual-energy X-ray absorptiometry (QDR-4500A; Hologic, Shelby Township, MI/Lunar iDXA, GE HealthCare, Chicago, IL). A negative pregnancy test was required for women of reproductive age. Baseline assessments of blood pressure, cfPWV, and brachial artery FMD were then performed, as described below, and repeated at 12 and 24 weeks of treatment. Participants underwent monitoring of renal function and potassium levels at 1, 4, 12, and 24 weeks. Additional monitoring was allowed per protocol at the discretion of the safety officer.

Assessment of blood pressure, cfPWV, and brachial artery FMD

After a period of rest (20 minutes), blood pressure measurements at the brachial artery were performed in duplicate using an automated blood pressure monitor (Welch Allyn Connex, Skaneateles Falls, New York, USA). cfPWV was measured using the SphygmoCor XCEL system (Cardiex, Sydney, Australia) with a cuff to assess aortic stiffness according to current recommendations,¹¹ and as described previously.¹² The SphygmoCor XCEL device is a well-validated, non-invasive measure of aortic stiffness that simultaneously captures aortic (tonometer) and femoral (cuff) pulse waves. Transit times were calculated using the foot-to-foot method, with wave feet identified using intersecting tangent algorithms. The cfPWV value, reported as m/s, was calculated by dividing the distance traveled by the pulse wave transit time. The structural and load-dependent arterial stiffness were calculated to determine the contribution of these mechanisms to any observed changes in cfPWV,^13–15 utilizing the following equation:

$$\mathit{Structural\; cfPWV}=\sqrt{\frac{A}{\mathit{Pref}}\bullet {\mathit{cfPWV}}^2-\raisebox{1ex}{$A$}\!\left/ \!\raisebox{-1ex}{$\rho $}\right.\bullet \ln \left(\frac{\mathit{Pref}}{A}\right)}$$

where A is an arbitrary reference mean arterial pressure for normotensive individuals (90 mmHg), Pref is the measured mean arterial pressure respective to each measurement, cfPWV is the unadjusted measured value, and $\rho$ is the density of blood (1050 kg/m³). Load-dependent stiffness was determined by taking the difference between the measured cfPWV and the calculated structural stiffness value.

For assessment of brachial artery FMD, a two-dimensional/Doppler ultrasound (GE Logiq P5) was used, following prior methods and published guidelines.¹⁶ Briefly, an 11-MHz linear array transducer was secured with a clamp and placed over the brachial artery. A rapid inflating cuff (Hokanson) was placed on the forearm ~5 cm distal to the antecubital fossa. In duplex mode, using a pulsed frequency of 5 MHz and corrected for a 60° insonation angle, simultaneous diameter and velocity signals were obtained. The sample volume was adjusted to encompass the entire vessel lumen without extending beyond the walls, and the cursor was set at mid-vessel. Real-time capture software (Elgato Video Capture, Elgato, CA) was used to record two minutes of ultrasound imaging, and then the cuff was inflated to a pressure of 250 mmHg for five minutes. Three minutes of ultrasound imaging were collected following cuff deflation. Specialized edge-detection software (Cardiovascular Suites 4, Quipu srl, Pisa, Italy) was used for offline analysis of recorded vascular variables. FMD percent change was calculated as follows:

$$\mathrm{F}\mathrm{M}\mathrm{D}(\%)=100\times \frac{D_{\mathit{peak}}-D_{\mathit{bl}}}{D_{\mathit{bl}}},$$

where $D_{\mathit{peak}}$ and $D_{\mathit{bl}}$ are the peak and baseline diameters reported in mm.

The shear rate was estimated as:

$$\mathrm{\gamma }=\frac{4\upsilon }{D},$$

where $\mathrm{\gamma }$ is the shear rate reported in s⁻¹, $\upsilon$ is the mean blood velocity in cm/s, and $D$ is the diameter in cm.

Other biochemical and metabolic markers

The homeostasis model assessment of insulin resistance (HOMA-IR) was calculated as: [fasting glucose (mg/dL) × fasting insulin (μU/mL)] / 405.¹⁷ The TyG index was calculated as: Ln[fasting triglycerides (mg/dL) × fasting glucose (mg/dL) / 2].¹⁸ Fasted plasma was sent to the University of Missouri Diabetes Diagnostic Laboratory for insulin analysis. Supine plasma aldosterone was sent to the Advanced Research and Diagnostic Laboratory at the University of Minnesota. Fasting plasma was collected similarly for analysis of lipid profile, glycated hemoglobin (A1C), and the basic metabolic panel (Quest Diagnostics Laboratories; Columbia, MO).

Statistical Analysis

Power calculations were conducted using G*Power (Ver. 3.1.9.7). Comparisons with placebo were planned a priori under the assumption of no systematic placebo-related change and were considered secondary. The study was primarily powered for within-group comparisons (power = 81 to 99% with effect size f = 0.10 to 0.38, correlation among repeated measures = 0.42 to 0.85) for the different variables examined. As shown in Figure 1, final enrollment was 88 in the amiloride group and 47 in the placebo group. Data are presented as mean with 95% confidence interval (CI). Linear mixed-effects models were used to evaluate changes in continuous outcome variables. Each outcome was analyzed in a separate model using the lmer() function from the lme4 package, with statistical inference for fixed effects performed using the lmerTest package (R Foundation for Statistical Computing, Vienna, Austria).^{19, 20} Fixed effects included time, group, and their interaction (time × group). A random intercept for participant ID was included to account for within-participant correlation due to repeated measurements. Models were estimated using restricted maximum likelihood (REML). The model for cfPWV was also adjusted for the change in SBP. Furthermore, in a secondary analysis, age and sex were included as covariates in all the reported linear mixed-effects models. Inclusion of age and sex as covariates did not change the direction or statistical significance of the primary findings. Missing data were handled by the default behavior of lmer function, which excludes observations containing NA values; no imputation was performed. Continuous data were visualized using GraphPad Prism (version 10.0) and are presented as mean values with 95% confidence intervals. Correlations between age and changes in cfPWV and SBP within each experimental group were assessed using Pearson or Spearman, as appropriate. A partial correlation was performed to examine the association between age and change in cfPWV while adjusting for potential confounding variables. Significance was set at P < 0.05. Statistical analysis was conducted in consultation with the Center for Applied Statistics and Data Analysis at the University of Missouri.

Figure 1. CONSORT flow diagram of participant enrollment, randomization, follow-up, and analysis.

Diagram depicts the number of participants assessed for eligibility, excluded before randomization, and randomized to amiloride or placebo. Shown are participant allocation, follow-up, exclusions after randomization, and the numbers included in the final analyses for each treatment group.

RESULTS

A total of 186 individuals were consented and screened. Of those, 137 were deemed eligible, randomized, and completed the baseline visit. Two subjects withdrew from the study before starting the study medication (Figure 1; CONSORT diagram). Table 1 summarizes subject characteristics, anthropometric and baseline blood profile parameters. The cohorťs mean age was 47 years, and 63.5% of participants were women, and of those 40% were postmenopausal (defined as at least 1 year away from last menstrual period). The racial and ethnic composition was as follows: 87.5% were White, 8.8% were Black, and 5.1% identified as Hispanic or Latino. The recruited cohort had elevated adiposity and insulin resistance indices. The baseline characteristics of the patients appeared balanced across the groups at time of randomization (Table 1). At baseline, 12.8% of participants in the placebo group were using antihypertensive medication, compared with 13.3% in the amiloride group. Additionally, 10.8% of participants in the placebo group were taking statins, compared with 5.0% in the amiloride group. A total of 116 participants completed the final visit (females, n=72; males, n=44, Figure 1).

Table 1.

Baseline subject characteristics and cardiometabolic measures prior to placebo or amiloride treatment.

	Placebo Mean [95% CI] (n=47, 28F)	Amiloride Mean [95% CI] (n=90, 59F)
Age (years)	47.60 [43.90,51.29]	46.30 [43.94,48.66]
Height (cm)	170.72 [167.96,173.49]	170.27 [168.41,172.12]
Body weight (kg)	103.97 [98.35,109.59]	99.76 [96.11,103.41]
Body mass index (kg/m2)	35.58 [33.99,37.17]	34.42 [33.24,35.60]
Body fat mass (%)	45.72 [43.96,47.47]	44.27 [42.67,45.88]
Systolic blood pressure (mmHg)	126.47 [121.40,131.54]	127.13 [124.40,129.87]
Diastolic blood pressure (mmHg)	78.09 [75.81,80.36]	78.68 [76.84,80.52]
Heart rate (bpm)	63.66 [60.78,66.54]	64.01 [61.96,66.06]
HOMA-IR	3.59 [2.89,4.28]	3.60 [3.17,4.04]
Estimated glomerular filtration rate (mL/min/1.73m2)	91.62 [86.79,96.46]	95.43 [92.47,98.39]
Potassium (mmol/L)	4.25 [4.18,4.32]	4.26 [4.20,4.32]
Glucose (mg/dL)	97.23 [94.03,100.44]	97.03 [95.20,98.87]
Aldosterone (ng/dL)	9.66 [7.96,11.37]	9.93 [8.66,11.20]
Total cholesterol (mg/dL)	197.91 [185.97,209.86]	207.70 [197.88,217.52]
HDL (mg/dL)	51.04 [47.53,54.56]	52.39 [49.22,55.56]
LDL (mg/dL)	122.43 [112.18,132.69]	127.66 [119.78,135.55]
Triglycerides (mg/dL)	136.30 [112.23,160.37]	151.84 [130.94,172.75]
Creatinine (mg/dL)	0.88 [0.82,0.93]	0.85 [0.82,0.87]
HbA1c (%)	5.40 [5.26,5.53]	5.28 [5.19,5.37]
Sodium (mmol/L)	138.60 [138.03,139.16]	138.82 [138.45,139.20]
Triglyceride-glucose (TyG) index	8.67 [8.53,8.81]	8.74 [8.62,8.86]
Carotid-femoral pulse wave velocity (m/s)	7.20 [6.77,7.63]	7.08 [6.82,7.34]
Brachial artery flow-mediated dilation (%)	5.22 [4.43,6.01]	5.51 [4.82,6.20]
Brachial artery shear rate AUC (x103 a.u.)	22.19 [18.61,25.77]	24.14 [20.98,27.30]

Data are presented as mean (95% confidence interval). Baseline characteristics were assessed before randomization to placebo or amiloride in participants who were overweight or obese and insulin resistant. Numbers in column headers indicate sample size and sex distribution. HOMA-IR, homeostatic model assessment of insulin resistance; HDL, high-density lipoprotein; LDL, low-density lipoprotein; HbA1c, glycated hemoglobin; TyG, triglyceride-glucose index.

There were no significant differences in aldosterone levels, cfPWV, SBP or DBP, heart rate, or brachial artery FMD between the amiloride and placebo groups at the start of the trial (Table 1). Baseline aldosterone levels did not differ between sexes (males: 10.38 ng/dL; 95% CI, 8.64 to 12.13; females: 9.52 ng/dL; 95% CI, 8.27 to 10.78). cfPWV decreased in the amiloride group from baseline by 0.24 m/s (95% CI −0.41 to −0.07) at 12 weeks (P < 0.05) and by 0.23 m/s (95% CI −0.39 to −0.07) at 24 weeks (P < 0.05) (Figure 2A). This reduction in cfPWV with amiloride was no longer statistically significant after adjusting for the change in SBP (Table 2). Congruently, there was a significant drop in the load-dependent cfPWV in the amiloride group at 24 weeks, while the structural-dependent cfPWV was not significantly altered (Table 2).

Figure 2. Effects of 24 weeks of amiloride or placebo and correlations with age.

Effects of amiloride or placebo on carotid-femoral pulse wave velocity (A), systolic blood pressure (B), potassium (C), and glucose (D). Mean changes from baseline were estimated using a linear mixed-effects model for repeated measures with fixed effects for treatment group and time. Error bars indicate 95% confidence intervals. The horizontal dotted line denotes no change from baseline. *P < 0.05 versus baseline within treatment group. Panels E and F display the correlations between age and changes in carotid-femoral pulse wave velocity and systolic blood pressure at 24 weeks within each experimental group. Each symbol represents an individual participant; squares indicate males and triangles indicate females. The horizontal dotted line denotes no change from baseline. Red lines represent linear regression fits with 95% confidence intervals. Correlation coefficients (Pearson or Spearman, as appropriate) and corresponding P values are shown within each panel.

Table 2.

Effects of 24 weeks of amiloride or placebo on anthropometric, metabolic, and vascular outcomes.

	Placebo Mean [95% CI] (n=37, 22F)	Amiloride Mean [95% CI] (n=79, 50F)
	Week 24	Week 24
Body weight (kg)	−0.14 [−1.37,1.10]	−0.53 [−1.37,0.32]
Body fat mass (%)	−0.12 [−0.64,0.40]	−0.14 [−0.50,0.21]
Diastolic blood pressure (mmHg)	0.68 [−1.41,2.76]	−2.35 [−3.79,−0.92]*
Heart rate (bpm)	1.49 [−1.31,4.28]	0.33 [−1.61,2.27]
Total cholesterol (mg/dL)	−15.22 [−24.15,−6.28]*	−11.01 [−17.13,−4.90]*
HDL (mg/dL)	−2.16 [−4.81,0.48]	−3.06 [−4.87,−1.25]*
LDL (mg/dL)	−8.48 [−15.58,−1.37]*	−7.16 [−11.99,−2.34]*
Triglycerides (mg/dL)	−21.11 [−43.70,1.48]	−12.35 [−27.81,3.11]
Triglyceride-glucose (TyG) index	−0.13 [−0.25,−0.01]*	−0.11 [−0.19,−0.02]*
HOMA-IR	−0.10 [−0.64,0.43]	−0.03 [−0.40,0.34]
Brachial artery flow-mediated dilation (% change)	0.89 [−0.37,2.15]	0.54 [−0.32,1.40]
Brachial artery shear rate AUC (x103 a.u.)	−3.20 [−8.18,1.79]	−3.20 [−6.60,0.19]
Estimated glomerular filtration rate (mL/min/1.73m2)	3.47 [0.28,6.66]*	−0.19 [−2.37,2.00]
Sodium (mmol/L)	−0.11 [−0.75,0.53]	−0.30 [−0.74,0.14]
Structural-dependent cfPWV (m/s)	−0.10 [−0.34,0.15]	−0.10 [−0.27,0.07]
Load-dependent cfPWV (m/s)	0.02 [−0.08,0.11]	−0.13 [−0.20,−0.07]*
cfPWV adjusted for change in SBP (m/s)	−0.08 [−0.33,0.17]	−0.18 [−0.37,0.01] (P = 0.064)

Mean changes from baseline were estimated using a linear mixed-effects model for repeated measures. The model for cfPWV was adjusted for the change in systolic blood pressure (SBP). Values in brackets indicate 95% confidence intervals. Numbers in column headers indicate sample size and sex distribution. HDL, high-density lipoprotein; LDL, low-density lipoprotein; HOMA-IR, homeostatic model assessment of insulin resistance; TyG, triglyceride-glucose index; eGFR, estimated glomerular filtration rate.

^*P < 0.05 versus baseline within treatment group.

Secondary analysis indicated that inclusion of sex and age into the model did not influence the effect of amiloride on cfPWV in that the effect remained statistically significant. However, older age was associated with a greater decrease in cfPWV (Figure 2E, r = −0.273, P = 0.018, n = 74). The correlation between age and change in cfPWV trended in the same direction but did not reach statistical significance when controlling for sex, BMI, and the change in SBP (r = −0.219, P = 0.067, n = 74). No significant changes were seen in the placebo group at either of the time points when compared with baseline (Figure 2A). For SBP, the amiloride group showed significant reductions of 2.96 mmHg at 12 weeks (95% CI −5.49 to −0.43; P < 0.05) and 5.59 mmHg at 24 weeks (95% CI −8.10 to −3.08; P < 0.05) relative to baseline (Figure 2B). As with cfPWV, sex and age did not influence the effect of amiloride on SBP, which remained statistically significant; however, older age was associated with a greater decrease in SBP (Figure 2F). Also, as with cfPWV, no significant changes in SBP were seen in the placebo group (Figure 2B). Treatment with amiloride was also associated with a lowering in DBP (Table 2), while no changes in heart rate were noted (Table 2). Brachial artery FMD was not significantly changed by the amiloride intervention (Table 2).

As expected for a potassium-sparing agent, the amiloride group increased potassium levels from baseline by 0.17 mmol/L at 12 weeks (95% CI 0.1 to 0.24; P < 0.05) and by 0.07 mmol/L at 24 weeks (95% CI 0.0 to 0.14; P < 0.05) (Figure 2C), and this effect was not influenced by sex and age. The eGFR decreased by 2.94 mL/min/1.73 m² (95% CI −5.14 to −0.74; P < 0.05) in the amiloride cohort at 12 weeks, but the reduction was no longer significant by 24 weeks (Table 2). On the other hand, in the placebo arm, eGFR increased significantly at 24 weeks (Table 2). Sex and age did not influence these changes. Glucose concentrations in the amiloride group decreased by 3.22 mg/dL at 12 weeks (95% CI −5.33 to −1.11; P < 0.05) and by 6.07 mg/dL at 24 weeks (95% CI −8.16 to −3.98; P < 0.05) relative to baseline (Figure 2D), with no influence of sex and age. The placebo group demonstrated smaller reductions that did not reach statistical significance at either 12 weeks or 24 weeks.

No severe adverse events were documented in either treatment group. Two participants (2.2%) in the amiloride group were withdrawn because of a >30% increase in serum creatinine from baseline; in accordance with the protocol, creatinine was monitored until normalization.

DISCUSSION

The primary findings of this randomized, placebo-controlled trial are that amiloride lowers blood pressure and arterial stiffness in individuals with obesity and insulin resistance, regardless of sex. Importantly, older individuals exhibited greater reductions in arterial stiffness and blood pressure following amiloride intervention. These changes were accompanied by increased serum potassium and decreased fasting blood glucose levels.

MR signaling increases ENAC activation in renal and vascular tissues. ENAC activity is not limited to sodium reabsorption by renal epithelium; our group and others have also demonstrated that ENAC activation in the vasculature leads to arterial stiffening.^{6, 7, 21, 22} We have also shown that overnutrition and insulin resistance are associated with increased presence of amiloride-sensitive sodium currents (i.e., ENAC activation) in endothelial cells, particularly in female mice.^{6, 7} Furthermore, ENAC blockade or deletion reduces arterial stiffness in preclinical models of obesity.^{6, 7} Despite preclinical evidence of greater ENAC activation in females with insulin resistance and obesity, in this clinical trial we document improvements in arterial stiffness and blood pressure with amiloride therapy in both males and females.

Arterial stiffening is not only associated with increased cardiovascular disease,²³ but it is also a hallmark of vascular aging.²⁴ We¹² and others^{25, 26} have previously reported that aging is characterized by elevated cfPWV. In the current investigation, amiloride treatment decreased cfPWV. Importantly, this was accompanied by a reduction in SBP at both 12 and 24 weeks. The magnitude of SBP lowering in the amiloride group, a 5.6 mmHg drop, has been associated with a significant decrease in cardiovascular disease risk.²⁷ Importantly, Paar et al. previously reported that aging is associated with increased ENAC expression in aortic endothelial cells and that ENAC activation may mediate blood pressure salt sensitivity in aging.²⁸ Our results support these preclinical findings, demonstrating that older individuals showed greater reductions in blood pressure and arterial stiffness after 24 weeks of amiloride treatment.

Hypertension and arterial stiffness are closely connected through reciprocal hemodynamic and vascular mechanisms. In this study, reductions in cfPWV with amiloride treatment were linked to simultaneous decreases in SBP, as shown by the reduced cfPWV effect after adjusting for changes in systolic pressure. This suggests that the improvement in arterial stiffness was mainly driven by load-dependent factors rather than structural changes in the arterial wall. Elevated blood pressure increases arterial wall stress, causing a functional rise in PWV through nonlinear pressure-diameter relationships, especially in elastic central arteries. Consistent with this, amiloride treatment significantly reduced the load-dependent component of cfPWV without affecting the structural-related stiffness. Accordingly, the current findings do not provide evidence of intrinsic modification of arterial stiffness but rather support a hemodynamic mechanism linking changes in SBP to improvements in PWV. These results underscore the close physiological coupling between blood pressure and cfPWV,^{29, 30} and highlight the importance of considering load-dependent effects when interpreting changes in arterial stiffness indices.

Obesity and insulin resistance are associated with impaired FMD. In our cohort, the mean brachial artery FMD at baseline visit was ~5% which is considered impaired.³¹ Amiloride intervention, however, did not improve brachial artery FMD in our study. In line with these findings, a previous investigation examining the effects of amiloride on nitric oxide metabolites and cyclic guanosine monophosphate production in patients with type 2 diabetes reported that, despite significant blood pressure lowering, amiloride treatment for 8 weeks did not alter plasma or urinary cGMP levels.³² On the other hand, a previous single-arm, non-placebo-controlled study by Bhagatwala et al. showed that 4 weeks of amiloride reduced blood pressure and carotid-radial PWV, but not cfPWV, in younger adults.³³ In that report, the authors provided evidence of improved brachial artery FMD with amiloride treatment. Notably, their study used a higher amiloride dose (10 mg daily) for a shorter duration in a significantly younger population with a lower BMI, all of which could have contributed to the differential results in FMD studies.

As previously stated, ENAC is a critical effector of the renin-angiotensin-aldosterone system, and its activation has been implicated in salt-sensitivity of blood pressure. Recently, ENAC activation in dendritic cells was shown to promote inflammation and salt-sensitive hypertension.³⁴ Furthermore, Jensen and colleagues reported that low-dose amiloride treatment in patients with diabetes resulted in lower levels of the proinflammatory cytokines TNF and IL-6.³⁵ Even though we did not examine inflammatory markers in our cohort, it is plausible that the improvement in arterial stiffness and blood pressure observed in our study was in part driven by a reduced proinflammatory milieu.

Our findings regarding plasma potassium warrant further discussion. As expected, amiloride therapy increased plasma potassium concentrations. Plasma potassium has been associated with improved vascular function through both vascular and renal effects.³⁶ Similarly, potassium concentrations are an independent predictor of diabetes incidence.³⁷ Therefore, it is reasonable to speculate that these changes in potassium levels may have mediated some of the positive effects observed in the amiloride group, independently of ENAC inhibition.

Amiloride has been previously reported to acutely impair insulin sensitivity (dose of 15 mg for 3 days).³⁸ However, in the present investigation, we did not observe changes in HOMA-IR or TyG index in response to amiloride. In addition, our finding of lower fasting glucose after 24 weeks of amiloride may be explained by the observed reduction in blood pressure.³⁹ In fact, a 5 mmHg reduction in blood pressure in pharmacological clinical trials has been associated with an 11% reduction in the risk of new-onset type 2 diabetes.⁴⁰

Of note, in our trial, no severe adverse effects were documented in the amiloride cohort. Nevertheless, we reported creatinine values >30% above baseline in two subjects, prompting discontinuation of the study per protocol. Both of those individuals exhibited normalization of creatinine after suspension of the study intervention.

Some limitations should be acknowledged. First, in this investigation, we did not assess blood pressure salt sensitivity. However, our cohort likely exhibited a high frequency of salt sensitivity, given that this phenomenon is more common in individuals with obesity and insulin resistance, and that salt-sensitive individuals would be expected to show greater responses to the intervention. Second, during the conduct of our clinical trial, the use of incretin-mimetic agents became increasingly common for the treatment of obesity and associated cardiometabolic abnormalities. Importantly, none of the participants in our cohort were receiving incretin-mimetic (GLP-1-based) therapies during their participation in the study. Therefore, it remains unknown if the arterial destiffening and blood pressure-lowering effects of amiloride that we report here would also be observed in patients receiving incretin mimetics. Third, this was a single-center study in which most participants were White. We recognize that multicenter trials with greater racial and ethnic diversity are needed to generalize our findings. Fourth, although greater reductions in arterial stiffness were observed among older participants, this finding should be interpreted appropriately. When controlling for sex, BMI, and the change in SBP, the correlation between age and change in cfPWV in the amiloride group was trending but no longer statistically significant. This indicates that age is not an independent determinant of amiloride-induced arterial destiffening and that the observed association is partially driven by other factors. Lastly, plasma concentrations of amiloride were not measured in the present investigation; thus, we cannot determine whether the observed improvements in vascular outcomes were mediated by renal ENAC blockade or by effects in non-epithelial tissues.

In conclusion, this randomized, placebo-controlled trial demonstrates that low-dose amiloride significantly reduces blood pressure and arterial stiffness in adults with obesity and insulin resistance, with older participants experiencing greater benefit. The improvement in arterial stiffness was largely dependent on the changes in blood pressure. This work highlights amiloride’s potential value as a therapeutic option for a population at increased cardiometabolic risk. The absence of significant adverse events further supports its feasibility in this setting.