Hypertension with primary aldosteronism is associated with increased carotid intima‐media thickness and endothelial dysfunction

Abstract Patients with primary aldosteronism induced hypertension are more likely to experience cardiovascular events compared to patients with essential hypertension. Primary aldosteronism may therefore have distinct adverse effects on cardiovascular structure and function, independent of hypertension. However, current data on such effects of primary aldosteronism are conflicting. The aim of the present study was to investigate the influence of primary aldosteronism on vascular structure and endothelial function, using intima‐media thickness as a vascular remodeling index and flow‐mediated dilation as a functional parameter. In total, 70 participants were recruited from patients with resistant hypertension. Twenty‐nine patients diagnosed with primary aldosteronism and 41 patients with essential hypertension were prospectively enrolled. Primary aldosteronism was due to aldosterone‐producing adenoma in 10 cases and due to idiopathic adrenal hyperplasia in 19 cases. All patients underwent ultrasound of the common carotid intima‐media thickness and flow‐mediated dilation of the brachial artery. Primary aldosteronism patients had significantly lower flow‐mediated dilation (3.3 [2.4‐7.4] % vs 14.7 [10.3‐19.9] %, P < 0.01) and significantly higher carotid intima‐media thickness (0.9 [0.7‐1.0] mm vs 0.8 [0.6‐0.9] mm, P = 0.02) compared to patients with essential hypertension. These differences remained significant after adjusting for age, sex, diabetes mellitus, 24‐hours systolic blood pressure, and smoking (P < 0.01). No differences in either outcome were observed between the adenoma and adrenal hyperplasia groups (both P > 0.05). Hypertensive patients with hyperaldosteronism appear to exhibit deteriorative effects on both vascular structure and function, independent of hypertension.

indicate that patients with PA are more likely to experience cardiovascular events and renal complications compared to patients with essential hypertension (EH). [2][3][4] These findings suggest that primary aldosteronism might have distinct adverse effects on cardiovascular structure and function, independent of hypertension.
Aldosterone overproduction leads to changes in vascular morphology, such as smooth muscle hypertrophy and collagen deposition within the extracellular matrix. 5,6 As a result, patients with PA have more pronounced fibrosis in the small resistance arteries. 7 Intima-media thickness (IMT), which can be assessed through ultrasound of the common carotid artery, is used as a marker of atherosclerosis. 8 Early studies reported that common carotid intima-media thickness (CC-IMT) is higher in patients with PA compared to patients with EH patients. 9,10 However, these studies were performed more than a decade ago. A more recent study reported no significant differences in baseline CC-IMT between subtypes of PA and EH patients. 11 Therefore, it remains debatable whether the subtypes of PA (ie, APA and IAH) exhibit different effects on cardiovascular structure.
Besides vascular structure, aldosterone also has profound effects on endothelial function. Aldosterone reduces endotheliumdependent vasodilation, either directly or through angiotensin II.
In experimental studies, it has been demonstrated that aldosterone prevents inducible nitric oxide (NO) synthase. 12 Flow-mediated dilation (FMD), which can be assessed with ultrasound of the brachial artery pre-and post-occlusion, is a widely used tool to directly and non-invasively evaluate endothelium-dependent vascular function. 13 Numerous studies have demonstrated that FMD is associated with the development of cardiovascular disease. 14,15 Although some studies reported that hypertensive patients with PA have lower FMD compared to patients with EH, 16,17 endothelial function paradoxically improves after 2 weeks of aldosterone excess. 18 Recently, it has been suggested that the detrimental effects of aldosterone overproduction are much less prominent in IAH than in APA, causing patients with PA due to IAH to have similar endothelial function as patients with EH. 19 Given the high prevalence (20%) of PA among patients with resistant hypertension, it is crucial to further elucidate the pro-atherogenic role of PA. 20 Therefore, the aim of the present study was to investigate the influence of PA and its subtypes on vascular structure and function, as assessed by CC-IMT and FMD.

| MATERIAL S AND ME THODS
Twenty-nine patients diagnosed with PA were prospectively enrolled. PA was due to APA in 10 (34%) cases and due to IAH in 19 (66%) cases. Forty-one patients with EH served as control group.
Participants were consecutively recruited from patients with resistant hypertension referred to the endocrinology or cardiology department of the Istanbul University of Medicine between April 2012 and November 2015. Resistant hypertension was defined as blood pressure above >140/90mm Hg in spite of optimal doses of 3 antihypertensive drugs, including at least one diuretic, or blood pressure requiring ≥4 antihypertensive drugs to control. 21 Patients with secondary causes of hypertension other than PA, such as Cushing's syndrome and renovascular hypertension, were excluded from the study.
Both PA and EH patients were subjected to a similar antihypertensive treatment regimen. In order not to interfere with the reninangiotensin-aldosterone system, pharmacological antihypertensive treatment was switched to monotherapy with a α-blocker (doxazosin) or duo-therapy with doxazosin and slow release verapamil. Oral potassium replacement was administrated to hypokalemic patients (<3.5 mEq/L) until a plasma level of >4.0 mmol/L was achieved.
The first study procedures were performed at least 14 days after adjustment of therapy. Demographics, medical history and current medication were collected. In addition, office blood pressure was measured twice on the right upper arm after a 15-minutes resting period using a manual sphygmomanometer. All patients also underwent 24-hours ambulatory blood pressure monitoring (SpaceLabs 90207; Medical). The study was conducted in accordance with the Declaration of Helsinki, and the study protocol was approved by the institutional review board of the Istanbul University, Istanbul Faculty of Medicine. All patients provided written informed consent for the study.

| Diagnosis of primary aldosteronism
Screening and diagnosis of PA were conducted according to the guidelines of the Endocrine Society. 22 As a first step, screening for PA was performed by sampling plasma aldosterone concentration (PAC) and plasma renin activity (PRA). Patients with PAC > 15 ng/ dL and plasma aldosterone/renin ratio (ARR) >20 were subjected to a saline infusion test or captopril challenge test to confirm the diagnosis of PA.
Saline infusion tests were regarded as positive if PAC was still >10 ng/dL following a 4-hours, 2L intravenous infusion of saline. In 5 patients with ARR > 20 and severe systolic hypertension, saline infusion was replaced by a captopril challenge test. Captopril challenge tests were regarded positive if PAC remained elevated and PRA remained suppressed 1 hour after receiving 25-50 mg of captopril orally. 22 In patients with confirmed PA, subtype classification was performed using computed tomography (CT) and/or magnetic resonance (MR) imaging of the adrenal glands.

| Echocardiography
All patients underwent standard echocardiography using two-di-

| Assessment of intima-media thickness
Ultrasound of the carotid artery was performed using a multifrequency (12 MHz) linear array transducer (GE Vivid 7.0, General Electric Vingmed Ultrasound), according to the latest Mannheim carotid intima-media thickness consensus statement. 24 Patients with previous carotid angioplasty or carotid endarterectomy and patients with carotid plaques more than 2.0 mm were excluded. All measurements were performed on high-resolution B-mode images. For CC-IMT measurements, patients were placed in the supine position with tilting of the head in a 45°C, opposite to the side of measurement. Longitudinal B-mode images of the distal segment (10-15 mm proximal to the carotid bifurcation) of the right common carotid artery were obtained in three consecutive regions of the common carotid artery at the far wall. Next, the distance between the bloodintima and media-adventitia interfaces was measured for all three regions. CC-IMT was thereafter computed by averaging the values of measurements.

| Assessment of flow-mediated dilation
Endothelium-dependent FMD was evaluated as described previously, using a multifrequency (12 MHz) linear array transducer. 25,26 Patients were placed in supine position, in a quiet, air-conditioned room (22-25°C), 15 minutes prior to the examination. High-resolution B-mode images of the brachial artery were obtained in the longitudinal plane, 3-5 cm proximal to the elbow of the right arm. Care was taken to maintain probe position during the investigation. Reactive hyperemia was induced by inflating a blood pressure cuff around the forearm for 4-5 minutes at 50 mm Hg above systolic blood pressure, followed by prompt deflation.
Brachial artery diameter was assessed by measuring the distance between the two lumen-intima interfaces at end-diastole. Three measurements were performed at baseline and post-occlusion.
Baseline and post-occlusive measurements were subsequently averaged. FMD was defined as the relative increase in brachial artery diameter ((diameter post-occlusion − basal diameter)/basal diameter) × 100)).
Echocardiography, assessment of intima-media thickness, and assessment of flow-mediated dilation were performed by consensus of two experienced cardiac ultrasonographers (AD and AE), who were blinded to clinical data (eg, cause of hypertension).

| Statistical analysis
Continuous variables are presented as mean ± standard deviation or median with interquartile range and were compared between two groups using the student t test for normally distributed data and Mann-Whitney test for non-parametric data. Categorical variables are expressed as frequency (percentage) and were compared between two groups using a chi-square test. Multiple group comparisons were performed by one-way analysis of variance (ANOVA), Kruskal-Wallis, or chi-square test according to variable type and distribution. Linear regression analysis was used to assess whether differences in FMD and CC-IMT remained significant after correcting for age, sex, diabetes mellitus, 24-hours systolic and diastolic blood pressure, and smoking. Association between continuous variables was quantified by Spearman's correlation.
Log transformation was applied to hormone levels in order to achieve parametric distribution. P-values <0.05 were considered statistically significant. Statistical analysis was done with the Statistical Package for Social Sciences software (IBM SPSS statistics 22 for Windows).

| RE SULTS
Clinical characteristics of the study cohort are listed in Table 1.
Patients with PA matched well with EH patients for age, gender, body mass index, and other cardiovascular risk factors. In addition, no differences in 24-hours average systolic and diastolic blood pressures were present, nor did the duration of hypertension differ between the groups. This was also demonstrated by similar IVST, PWT, and diastolic dysfunction. Furthermore, there were no significant differences for chronic antihypertensive medication between patients with PA and EH. With regard to kidney function, patients with PA had a higher prevalence of proteinuria (41% vs 17%, P = 0.02) and As expected, hypokalemia was more frequently present in patients with PA compared to patients with EH (48% vs 2.4%, P < 0.01).
Within the group of PA, hypokalemia was more frequently observed in patients with APA than patients with IAH (80% vs 31%, P = 0.01).

| Hormonal levels
As expected, patients with PA demonstrated higher PAC and ARR values and lower PRA values compared to patients with EH (P < 0.01). In addition, no significant difference was observed for PAC, ARR, and PRA between subtypes of PA (P > 0.05).

| D ISCUSS I ON
The present study is the first to perform combined assessment of vascular structure and function in patients with primary aldosteronism

| Primary aldosteronism as an independent risk factor for future cardiovascular events
Numerous retrospective and prospective studies indicated that patients with PA are at higher risk for cardiovascular events compared to patients with EH. 3

| Subtype classification of primary aldosteronism
Therapeutic approaches in PA significantly and distinctly differ depending on the lateralization finding; thus, application of an accurate technique for subtype classification is great of importance.
The recent guidelines of the Endocrine Society advocated AVS as the gold standard to distinguish between subtypes of PA suggesting use of CT as a stand-alone technique only for young patients (<35 years) with spontaneous hypokalemia and marked aldosterone excess. 22 However, CT may underdiagnose adenomas below 10 mm in diameter due to insufficient spatial resolution, whereas successive AVS demonstrates high sensitivity and specificity (95% and 100%, respectively) to distinct lateralization. 34

| Vascular structure
Experimental studies have shown that excess of aldosterone stimulates vascular smooth cell hypertrophy, adventitial cell migration, and collagen accumulation in the extracellular matrix. [5][6][7] These structural changes ultimately result in vascular remodeling which might explain the increased cardiovascular events in patients with PA. IMT is a valid and useful tool to investigate vascular structural changes.
In several studies, a remarkable relationship was demonstrated between IMT and cardiovascular risk factors. 37

| Endothelial function
Aldosterone also exhibits adverse effects on endothelial function.
Aldosterone was reported to directly limit NO production by decreasing NO synthase activity and increasing superoxide anion generation in endothelial cells. 41 However, aldosterone levels were also found to be higher in patients with APA compared to patients with IAH. The attenuated post-occlusive response can therefore not solely be attributed to subtype.
In the present study, aldosterone levels were similar between patients with mass and hyperplasia and no differences in FMD were observed between the subtypes.

| Hormonal data
In the present study, no correlation was found between IMT and hormone levels (PAC, PRA, ARR). These results are in accordance with literature. 9,10 Furthermore, a previous study found no correlation between PAC and regression of IMT after treatment of PA. 11 These apparently inconsistent findings can be explained by several reasons. Firstly, blood samples were collected 2 weeks after discontinuation of medical therapy, and therefore, the hormonal levels during the chronic phase might differ. Secondly, additional factors such as density of aldosterone receptors may vary depending on the region and can be determinative in the occurrence and severity of the vascular damage, as well.
In the present study, FMD correlated significantly with ARR and also PRA, whereas no correlation was detected between FMD and PAC. The association of FMD with both ARR and PRA remained significant after correcting for blood pressure. Although some studies have reported that FMD is associated with hormone levels, this is still a matter of debate. 16,17,44 Of note, ARR provides a more robust measure than PAC and PRA for evaluation of aldosterone overproduction, since PAC and PRA are affected by several confounders such as dietary sodium intake and posture. 45 Consequently, ARR might be more closely associated with parameters of cardiovascular structure and function than PAC and PRA. However, CC-IMT and FMD represent generalized atherosclerosis and other cardiovascular risk factors can still confound the relationship with ARR.

| Implications and future perspectives
These data clearly confirm that exposure to inappropriate aldosterone overproduction for sodium levels is associated with increased cardiovascular risk by altering both cardiovascular structure and

| S TUDY LIMITATI ON S
This study has some important limitations. First, as previously mentioned, the number of patients with PA is relatively small, particularly for APA (10), which may have precluded the comparison between PA subtypes, as well as differences in clinical and biochemical parameters. Second, adrenal vein sampling (AVS), which is the preferred method for subtype classification of PA as recommended by the latest guideline of the Endocrine Society, was attempted in only four patients. 22 Because of the low success rate of the procedure, the procedure was not performed in the remainder of patients. However, subtyping was performed using CT and/or MR imaging for all the patients with confirmed PA. Nevertheless, the results regarding comparison between subtypes of PA should be interpreted cautiously, since the distinction of subgroups is not definitive. Finally, a true control group is missing. However, the rationale of the study was able to assess the severity of atherosclerotic features in the patients with PA in comparison to patients with EH.

| CON CLUS IONS
Primary aldosteronism appears to exhibit deteriorative effects on both vascular structure and function, as determined with carotid intima-media thickness and flow-mediated dilation, independent of hypertension.

ACK N OWLED G M ENTS
None.

CO N FLI C T O F I NTE R E S T
This study was initiated by the Istanbul University, Istanbul Faculty of Medicine, and the authors report no specific funding in relation to this research and no conflicts of interest to disclose.