Sex‐related associations of high‐density lipoprotein cholesterol with aortic stiffness and endothelial glycocalyx integrity in treated hypertensive patients

Abstract

Smoking, a well‐recognized major cardiovascular (CV) risk factor, impairs endothelial function and increases aortic stiffness which indicates subclinical organ damage in hypertensive patients. Loss of endothelial glycocalyx (EG) integrity, as part of the endothelium, represents endothelial dysfunction. The authors aimed to investigate the role of increased HDL cholesterol levels (HDL‐C), which usually are considered protective against CV disease, in aortic stiffness and endothelial integrity in middle‐aged treated hypertensive patients regarding smoking habit. The authors studied 193 treated hypertensive patients ≥40 years (mean age = 61±11 years, 58% females), divided in four groups regarding sex and smoking. Increased perfusion boundary region of the 5‐9 μm diameter sublingual arterial microvessels (PBR_5‐9) was measured as a noninvasive accurate index of reduced EG thickness. Aortic stiffness was estimated by carotid‐femoral pulse wave velocity (PWV). In the whole population, an inverse weak relationship was found between HDL‐C and PWV (r = −.15, P = .03) and PBR_5‐9 (ρ = −.15, P = .03). Moreover, HDL‐C was negatively related to PBR_5‐9 in males (r = −.29, P = .008) either smokers (r = −.35, P < .05) or non‐smokers (r = −.27, P < .05) and PWV in female non‐smokers (r = −.28, P = .009). In a multiple linear regression analysis, using age, weight, smoking, HDL‐C, and LDL‐C as independent variables, we found that HDL‐C independently predicts PWV in the whole population (β = −.14, P = .02) and PBR_5‐9 in male hypertensive patients (β = −.28, P = .01). Higher HDL‐C levels are associated with reduced aortic stiffness in hypertensive patients, while they protect EG and subsequently endothelial function in middle‐aged, treated hypertensive male patients (either smokers or not).

1. INTRODUCTION

Arterial hypertension affects more than ¼ of the population worldwide representing one of the most important risk factors for cardiovascular disease (CVD). The complications of hypertension disease vary from subclinical organ damage (hypertension mediated organ damage, HMOD) to coronary artery disease, stroke, heart failure, and chronic kidney disease, major causes of morbidity and mortality. ¹ HOMD (left ventricular hypertrophy, reduced coronary flow reserve, increased carotid intima‐media thickness, microalbuminuria, aortic stiffness, and endothelial dysfunction) constitutes an intermediate stage of cardiovascular (CV) continuum, and it should be cautiously investigated in hypertensive patients in order to estimate individual CV risk and design the appropriate therapeutic plan.

High‐density lipoprotein cholesterol (HDL‐C) is another CV risk factor. Elevated HDL‐C plasma levels seem to exert protective role against CVD, ² , ³ establishing an inverse relationship with coronary events, ischemic stroke, and carotid atherosclerosis even after adjustment for lipid and non‐lipid risk factors in populations without known CVD. ⁴ However, the failure of recent trials to decrease CV events despite raising HDL‐C levels with various treatment methods has called into question epidemiological data from previous years. ⁵

The endothelial glycocalyx (EG) is the most luminal layer of the blood vessel, consisting of glycoproteins and proteoglycans. Certain morbidities, arterial hypertension included, lead to enhanced vascular sensitivity to prothrombotic, vasoactive, and atherogenic stimuli causing EG degradation even in early stages, which is associated with increased morbidity and mortality. ⁶ EG function is assessed through measurement of perfused boundary region (PBR) of the sublingual microvessels ranged 5‐25 μm. ⁷ In a previous study, we proved that HDL‐C ranged between 71 and 101 mg/dL might moderately protect EG and subsequently endothelial function in hypertensive patients, older than 50 years. ⁸ In this study, we aimed to investigate the role of increased HDL‐C levels, which usually are considered protective against CVD, in aortic stiffness and endothelial integrity in middle‐aged treated hypertensive patients regarding smoking habit.

2. MATERIAL AND METHODS

2.1. Study population

We studied a cohort of 193 consecutive patients aged ≥40 years with essential hypertension under treatment but no other comorbidities (except diabetes mellitus and/or hyperlipidemia). These patients visited our University hospital outpatient Hypertension Clinic as new referrals or scheduled follow‐ups between November 2015 and November 2019. All patients were subjected to the following examinations: (a) office blood pressure (BP) measurements in the hypertension outpatient clinic; (b) blood sampling for routine blood chemistry examination; (c) fasting lipid profile measurement in a core laboratory (total cholesterol, triglycerides, HDL‐C, and LDL); (d) standard 12‐lead electrocardiogram; (e) carotid‐femoral pulse wave velocity measurement (PWV) in order to evaluate aortic stiffness; and (f) PBR of the sublingual arterial microvessels (diameter ranged 5‐25 μm) measurement in order to evaluate EG integrity.

Patients were advised to continue their antihypertensive, antidiabetic, and hypolipidemic treatment on the day of the examination since any short‐term interruption of the treatment should not have an effect on our results.

Patients with untreated hypertension, secondary hypertension, congestive heart failure, previous myocardial infarction, stroke, cardiac valve diseases, history of coronary artery intervention and/or by‐pass grafting, atrial fibrillation, renal insufficiency (estimated glomerular filtration rate, eGFR ≤ 60 mL/min/1.73 m² using the formula CKD‐EPI), overt proteinuria (urine albumin levels ≥300 mg/24 h), non‐ CVDs, pregnancy, and hormone replacement treatment (estrogens and/or progesterone preparations) were excluded from the study in order to study a homogenous hypertensive group.

Informed consent had been obtained during the initial visit of the study which was approved by the ethical committee of the hospital.

2.2. Diagnostic work‐up

The protocol of the study has been described with details in a previous study. ⁸ However, a short description follows:

2.2.1. Office BP measurement

Office BP was measured in the hospital outpatient clinic, approximately in the same morning hour of the day, by the same cardiologist with an approved BP monitor after the patients had rested for a period of 5‐10 minutes in the sitting position. Two measurements were taken at 1‐minute intervals, and the average was used to define office systolic (SBP) and diastolic (DBP) blood pressure.

2.2.2. Carotid‐femoral pulse wave velocity (PWV)

Aortic stiffness was estimated by an automatic carotid‐femoral PWV measurement using a Complior SP (Artech Medical), a computerized device that permits automatic calculation of PWV. The same examiner, who was blinded to the patient's history, performed all measurements. Patients were advised to avoid smoking or coffee at least for 2 hours before examination. PWV ≤ 10 m/s was considered as normal. ¹ , ⁹

2.2.3. Endothelial glycocalyx (EG)

We measured the PBR of the sublingual arterial microvessels (diameter ranged 5‐25 μm) using Sidestream Darkfield imaging (Microscan, Glycocheck, Microvascular Health Solutions Inc.) which is proposed as a valid method to assess endothelial integrity by the European Society of Cardiology Working Group on Peripheral Circulation. ⁸ , ¹⁰ , ¹¹ , ¹²

2.2.4. Lipid profile

All patients were subjected in fasting lipid profile measurement in a core laboratory (total cholesterol, triglycerides, HDL‐C, and LDL‐C). Both lipid profile estimation and PBR measurement were performed the same day of each patient's evaluation.

Subsequently, hypertensive patients were divided regarding sex and smoking habit in four groups: male smokers (n = 31), male non‐smokers (n = 51), female smokers (n = 22), and female non‐smokers (n = 89).

2.3. Statistical analysis

Variables were tested by the Shapiro‐Wilk test to assess the normality of distribution. In the whole, smokers and non‐smokers’ population the following parameters were not normally distributed: weight, BMI, triglycerides, central SBP, HDL‐C, PWV, and PBR_5‐9. These non‐normally distributed parameters are expressed as median value plus 25%‐75% interquartile range (25‐75 IQR). All other normally distributed variables are expressed as mean ± standard deviation. Categorical variables are expressed as absolute values and percentages. Paired Student's t test was used in order to compare numeric differences regarding normally distributed variables within groups. Mann‐Whitney U test was used for non‐normally distributed parameters or categorical variables. Pearson's correlation was used to identify the relations between PBR_5‐25, PWV, and HDL‐C levels in the whole population as well as several study subgroups. Statistical analysis was performed on a SPSS 21 version (SPSS Inc), while P value <.05 was the level of significance.

Multiple linear backward regression analysis, using stepwise method, was performed in order to explore independent relations of HDL‐C with PWV and PBR_5‐9 measurements in the whole population as well as study groups (males, females). Age, weight, smoking habit, LDL‐C, and HDL‐C were forced into the regression analysis models as independent variables since they play a significant role in arterial stiffness as well as endothelial function. Since age, weight, smoking habit, LDL‐C, and HDL‐C are correlated; we used weighted least squares (WLS) instead of ordinary least squares (OLS) in order to reveal more unbiased results. Finally, we created two models in order to study the independent association between HDL‐C and PWV (Model A) and HDL‐C and PBR_5‐9 (Model B) in the whole population as well as in males and females separately.

3. RESULTS

Demographic and clinical characteristics of total population and studied groups (smokers, non‐smokers) are listed in Table 1. The whole population (mean age ± SD=61 ± 11 years, 58% females) was mostly consisted of non‐smoker (72%) but almost obese hypertensive patients (median BMI = 29 kg/m²) with a small incidence of diabetes mellitus (14%). An inverse but weak relationship was found between HDL‐C levels and PWV (r = −.15, P = .03) as well as PBR_5‐9 (ρ = −.15, P = .03).

TABLE 1.

Study population demographic and clinical characteristics

	Total population	Smokers				Non‐smokers				P
		All	Males	Females	p1	All	Males	Females	p2
Characteristics
N	193	53	31 (58%)	22 (42%)		140	51 (36%)	89 (64%)
Age (years)	61 ± 11	57 ± 9	56 ± 9	59 ± 10	0.22	63 ± 12	62 ± 11	63 ± 12	0.42	.001
Weight (kg)	78 (70‐90)	82 (73‐96)	92 (83‐101)	70 (60‐78)	0.24	76 (68‐87)	86 (75‐98)	72 (65‐82)	0.21	.04
BMI (kg/m2)	29 (26‐32)	28 (26‐31)	29 (27‐31)	27 (23‐32)	0.07	29 (26‐33)	28 (26‐33)	29 (26‐33)	0.69	.33
Current smokers (%)	53 (28%)	53 (100%)	31 (100%)	22 (100%)	‐	0 (0%)	0 (0%)	0 (0%)	‐	‐
Diabetes mellitus (%)	26 (14%)	7 (13%)	4 (13%)	3 (14%)	0.94	19 (14%)	9 (18%)	10 (11%)	0.31	.94
Total cholesterol (mg/dL)	196 ± 39	201 ± 45	202 ± 44	199 ± 46	0.81	194 ± 36	192 ± 43	196 ± 32	0.56	.29
Triglycerides (mg/dL)	99 (73‐149)	123 (76‐164)	140 (79‐245)	108 (71‐142)	0.13	98 (72‐141)	112 (77‐144)	93 (67‐138)	0.08	.08
LDL‐C (mg/dL)	124 ± 33	128 ± 36	133 ± 36	122 ± 37	0.27	122 ± 32	129 ± 36	118 ± 29	0.06	.23
HDL‐C (mg/dL)	62 (42‐73)	46 (35‐69)	37 (32‐54)	68 (51‐78)	<0.001	65 (47‐75)	48 (37‐64)	70 (60‐77)	<0.001	.003
Treatment with statins (%)	84 (44%)	26 (49%)	15 (48%)	11 (50%)	0.91	58 (41%)	18 (35%)	40 (45%)	0.26	.34
Office SBP (mm Hg)	144 ± 18	139 ± 14	138 ± 13	140 ± 16	0.61	146 ± 19	139 ± 15	150 ± 19	0.001	.006
Office DBP (mm Hg)	87 ± 8	86 ± 8	87 ± 6	84 ± 8	0.18	88 ± 10	87 ± 8	88 ± 11	0.79	.31
Central SBP (mm Hg)	133 (123‐138)	127 (121‐138)	127 (119‐136)	128 (121‐139)	0.75	134 (124‐151)	127 (119‐145)	139 (129‐154)	0.002	.01
PWV (m/s)	11.5 (10‐13)	11.3 (10‐13)	11.2 (10‐13)	11.4 (10‐12)	0.77	11.8 (10‐13)	11.6 (10‐13)	11.9 (10‐13)	0.87	.32
PBR 5‐25 (μm)	2.1 ± 0.2	2.1 ± 0.3	2 ± 0.3	2.3 ± 0.2	0.001	2.1 ± 0.2	2.1 ± 0.2	2.2 ± 0.2	0.02	.39
PBR 5‐9 (μm)	1.2 (1.1‐1.3)	1.2 (1.1‐1.3)	1.2 (1.1‐1.2)	1.2 (1.1‐1.3)	0.29	1.2 (1.1‐1.3)	1.2 (1.1‐1.3)	1.2 (1.1‐1.3)	0.17	.94
PBR 10‐19 (μm)	2.3 ± 0.3	2.2 ± 0.4	2.1 ± 0.3	2.4 ± 0.3	0.003	2.3 ± 0.3	2.2 ± 0.3	2.3 ± 0.3	0.02	.53
PBR 20‐25 (μm)	2.7 ± 0.4	2.7 ± 0.4	2.4 ± 0.4	2.9 ± 0.3	<0.001	2.7 ± 0.4	2.6 ± 0.4	2.7 ± 0.4	0.05	.30

Data are presented as mean ± SD for normal distributed parameters and as median + 25‐75 Interquartile Range (IQ) for non‐normal distributed parameters.

Italic values correspond to statistical significance.

Abbreviations: BMI, Body mass index; DBP, diastolic blood pressure; HDL‐C, high‐density lipoprotein cholesterol; LDL‐C, low‐density lipoprotein cholesterol; p refers to differences between smoker and non‐smoker hypertensive patients; p₁ refers to difference between males and females smoker hypertensive patients; p₂ refers to difference between males and females non‐smoker hypertensive patients; PBR, perfusion boundary region; PP, pulse pressure; PVW, pulse wave velocity; SBP, systolic blood pressure.

Patients in both studied groups (smokers, non‐smokers) had similar BMI, incidence of diabetes mellitus, total cholesterol, triglycerides, and LDL‐C levels as well as percentages of treated patients with statins, office diastolic and central systolic BP, aortic stiffness (PWV), and endothelial function (PBR_5‐25, PBR_5‐9, PBR_10‐19, PBR_20‐25). We found that non‐smokers (mean age = 63±12 years, 36% males) had median HDL‐C = 65 mg/dL (IQ = 47‐75) and office SBP = 139±14 mm Hg, while smokers (mean age = 57±9 years, 58% males) had median HDL‐C = 46 mg/dL (IQ = 35‐69) and office SBP = 146±19 mm Hg.

3.1. Hypertensive smoker patients

We found that smoker females compared to smoker males had similar age, weight, BMI, incidence of diabetes mellitus, total cholesterol, triglycerides and LDL‐C levels (but increased HDL‐C, P < .001), office SBP and DBP, central SBP and PWV. However, endothelial function (PBR_5‐25, PBR_10‐19, PBR_20‐25) was more impaired in smoker females compared to smoker males, except PBR_5‐9 which was similar.

We performed Pearson's correlation in the whole population of smoker hypertensive patients (Table 2), and we found that HDL‐C levels were negatively related to PBR_5‐9 (r = −.24, P = .08). Regarding sex distribution, in the group of smoker male patients, HDL levels were negatively related to PBR_5‐25 (r = −.38, P = .03), PBR_5‐9 (r = −.35, P = .05), and PBR_10‐19 (r = −.41, P = .02) while no relationship was revealed with PWV. On the other hand, in the group of smoker female patients, no significant relationship was revealed between HDL levels and aortic stiffness or endothelial function indices.

TABLE 2.

Correlations between HDL, PBR, and PWV in several study groups

Parameters	H D L ‐ C
	Total population			Smokers			Non‐smokers
	All (n = 193)	Males (n = 82)	Females (n = 111)	All (n = 53)	Males (n = 31)	Females (n = 22)	All (n = 140)	Males (n = 51)	Females (n = 89)
PWV (m/s)	r = −.15, P = .03	NS	r = −.23, P = .01	NS	NS	NS	r = −.21, P = .01	NS	r = −.28, P = .009
PBR 5‐25 (μm)	NS	NS	NS	NS	r = −.38, P = .03	NS	NS	NS	NS
PBR 5‐9 (μm)	ρ = −.15, P = .03	r = −.29, P = .008	NS	r = −.24, P = .08	r = −.35, P = .05	NS	NS	r = −.27, P = .05	NS
PBR 10‐19 (μm)	NS	NS	NS	NS	r = −.41, P = .02	NS	NS	NS	NS
PBR 20‐25 (μm)	NS	NS	NS	NS	NS	NS	NS	NS	NS

Abbreviations: HDL‐C, high‐density lipoprotein cholesterol; NS, no significant correlation; PBR, perfusion boundary region; PVW, pulse wave velocity.

3.2. Hypertensive non‐smoker patients

We found that females compared to males had similar age, weight, BMI, incidence of diabetes mellitus, total cholesterol, triglycerides and LDL‐C levels (but increased HDL‐C, P < .001), office DBP, PBR_5‐9, and PWV. However, office and central SBP as well as endothelial function (PBR_5‐25, PBR_10‐19, and PBR _20‐25) were more impaired in females compared to males, except PBR_5‐9 which was similar.

After Pearson's correlation in the whole population of non‐smoker hypertensive patients (Table 2), we found an inverse correlation between HDL‐C and PWV (r = −.21, P = .01). Regarding sex distribution, in the group of male patients, HDL‐C levels were negatively related to PBR_5‐9 (r = −.27, P = .05) while no significant linear relationship has been revealed regarding PWV. On the other hand, in the group of female patients, a significant correlation was revealed between HDL‐C levels and PWV (r = −.28, P = .009).

3.3. Multiple linear regression analysis

We performed multiple linear regression analysis using stepwise method. Age, weight, smoking habit, and LDL‐C levels were inserted as independent variables in order to investigate any independent associations between HDL‐C and PWV (Model A) as well as PBR_5‐9 (Model B) in the whole population, in males and females (Table 3). Since age, weight, smoking habit, LDL‐C, and HDL‐C are correlated, we used WLS instead of OLS in order to reveal more unbiased results.

TABLE 3.

Multiple linear regression analysis regarding independent associations of aortic stiffness and endothelial function

Independent variables	Model A (Dependent variable PWV)			Model B (Dependent variable PBR5‐9)
	All (n = 193)	Males (n = 82)	Females (n = 111)	All (n = 193)	Males (n = 82)	Females (n = 111)
Age (years)	β = .43, P < .001	β = .42, P < .001	β = .46, P < .001	β = .16, P = .02	‐	‐
Weight (kg)	‐	‐	‐	β = .15, P = .03	‐	‐
Smoking habit	‐	‐	‐	‐	‐	‐
LDL (mg/dL)	‐	‐	‐	‐	‐	‐
HDL (mg/dL)	β = −.14, P = .02	‐	‐	‐	β = −.28, P = .01	‐

Abbreviations: HDL‐C, high‐density lipoprotein cholesterol; LDL‐C, low‐density lipoprotein cholesterol; PBR_5‐9, perfusion boundary region; PVW, pulse wave velocity.

In Model A, we found that a. in the whole population, HDL and age independently predict PWV (β = −.14, P = .02 and β = .43, P < .001, respectively), b. in both males and females, age independently predicts PWV (β = .42, P < .001 and β = .46, P < .001, respectively).

In Model B, we found that a. in the whole population, age and weight independently predict PBR_5‐9 (β = .16, P = .02 and β = .15, P = .03, respectively) and b. in males, HDL independently predicts PBR_5‐9 (β = −.28, P = .01).

4. DISCUSSION

In the present cross‐sectional study, we tried to investigate the sex‐related vascular effects of HDL‐C in aortic stiffness and endothelial integrity of treated hypertensive patients aged ≥40 years. The primary endpoint of the study is that HDL‐C moderately and inversely predicts a. aortic stiffness (PWV) in the whole hypertensive population and b. endothelial function (PBR_5‐9), a negative index of EG integrity, in treated male hypertensive patients independently of their smoking habit. The prognostic value of increased HDL‐C remains a well‐established knowledge contributing to CV protection through a variety of functions (reverse cholesterol transport, maintenance of endothelial cell homeostasis, and antioxidant properties). ¹³ , ¹⁴ On the other hand, lower levels of HDL‐C can be pro‐inflammatory, contributing to disease progression in patients with atherosclerosis. ¹⁵ The pharmacological efforts to increase HDL‐C levels did not show any benefits in recent clinical trials but increased mortality in one study as well. ¹⁶ , ¹⁷ Furthermore, extremely high HDL‐C levels do not seem to be always atheroprotective. ¹⁸ This could be due to genetic mutations in HDL‐C metabolism, so even in increased HDL‐C levels, the HDL‐C particles cannot function properly and subsequently increase CV risk. ¹⁸

Although EG dysfunction might define prognosis in various CV diseases, data regarding hypertension disease are relatively insufficient. ¹⁹ EG dysfunction represents an index of HMOD. ¹ HDL‐C plays a vital role to endothelial cell homeostasis through nitric oxide release and subsequent promotion of vasorelaxation, inhibition of cell adhesion and pro‐inflammatory molecules expression and endothelial layer repair. ²⁰ , ²¹

Numerous methods for the evaluation of the endothelial function have been used in the past (flow‐mediated vasodilation, carotid intima‐media thickness, and serum biomarkers). ⁷ In our study, we chose to use a new technique, the sideview darkfield imaging in order to visualize and evaluate the integrity of EG, a significant part of endothelium. We revealed a moderate and inverse association between HDL‐C levels and PBR_5‐9 in male hypertensive patients (noticeably independent of smoking habit). Thus, we might infer that the role of increased HDL‐C levels accompanied by improved endothelial function, as a favorable CV risk factor, is more significant in males compared to females. On the other hand, females usually have higher HDL‐C levels compared to males, ²² , ²³ , ²⁴ , ²⁵ , ²⁶ as it has also been proven in our study. However, we did not prove any relationship between HDL‐C and EG in females which in turn enhances our conclusions regarding the significant role of increased HDL‐C in hypertensive males regarding EG function and the subsequent CV risk reduction. Nonetheless, the moderate relationship between HDL‐C and EG in males probably reflects the existence of other factors that affect EG function or even the presence of dysfunctional HDL particles in extremely elevated HDL‐C ⁸ (Figure 1) .

FIGURE 1.

The independent association between HDL‐C levels and PBR_5‐9 in male hypertensive patients

Cigarette smoking is a major risk factor for CVD, exerting its effects by increasing arterial stiffness, leading to endothelial dysfunction, and accentuating oxidative stress. ²⁷ , ²⁸ In a previous study, we just found a trend toward an inverse relationship between HDL‐C and PBR_5‐9 in a small group of smoker hypertensives under treatment. ⁸ In the present study, an inverse association of HDL‐C and PBR_5‐9 was revealed in male hypertensive patients while we have not been able to prove any linear dependence of the following factors of age, weight, LDL‐C levels, and smoking habit to PBR_5‐9. Taking this result into consideration, we can hypothesize that increased HDL levels might possess a crucial independent role in EG integrity in males, smokers or not.

Endothelial glycocalyx diameter and subsequently integrity is inversely related to PBR. ⁷ We found that PBR (PBR_5‐25, PBR_10‐19, PBR_20‐25) was increased in women (either smokers or non‐smokers) compared to men. However, there was no significant difference regarding PBR_5‐9 within sex groups. In a multi‐ethnic community‐based sample, the highest PBR was most strongly associated with older age, female sex, higher DBP, lower BMI, and diabetes. Subsequently, the association of highest PBR with female sex does not fit the hypothesis that a small glycocalyx size reflects higher CV risk. ²⁹ Others researchers have suggested that glycocalyx size as estimated by the PBR should be standardized for hematocrit, since an association between higher hematocrit and lower PBR has been observed and that may in part explain the observed association with sex and age. ³⁰ , ³¹ In coronary artery disease, both males and females have significantly impaired sublingual microvascular perfusion compared to controls. However, women, but not men, exhibited a significantly larger PBR, indicting a reduction in glycocalyx barrier function suggesting sex dependency. ³²

Pulse wave velocity is the most validated marker of aortic stiffness and an independent CV risk factor. ³³ Although smoking is associated with an increased aortic stiffness in smokers regardless of gender, both arterial stiffness and central blood pressure increase in female smokers after less heavy smoking compared to male smokers. ³⁴ The differential clinical characteristics of the development of arterial stiffness between men and women indicate the involvement of sex‐specific mechanisms. ³⁵ The association of HDL with arterial stiffness has been investigated both in normotensives and hypertensives patients. It seems that triglyceride/HDL‐C ratio was directly and independently associated either with radial pulse wave analysis in a Finnish population or with brachial‐ankle PWV (baPWV) in a Japanese cohort, while HDL‐C was inversely related to baPWV in a Chinese population. ³⁶ , ³⁷ , ³⁸ Similarly, in our study, we found a moderate inverse association between HDL‐C and carotid‐femoral PWV in the whole hypertensive population, while we did not prove any linear dependence of the following factors of age, weight, LDL‐C levels, and smoking habit to PWV_.

4.1. Study limitations

The cross‐sectional design of our study and the moderate number of treated hypertensive patients, overall and in each study group, prevented us from inferring definite cause‐effect relationships as well as to generalize our results in untreated and treated hypertensive patients as well as in groups of treated hypertensives under different classes of antihypertensive drugs. Additionally, our study was just based on a single HDL‐C assessment while we were not able to access the size of HDL‐C particles, especially the large ones, which might not be protective against atherosclerosis and endothelial function. However, strength of our study is the performance of HDL‐C measurements in a core laboratory which are referred to a relatively homogenous hypertensive group since all patients were ≥40 years with no other comorbidities (only 14% had diabetes mellitus and 44% had hyperlipidemia under treatment with statins). Since HDL‐C levels ranged between 23 and 118 mg/dL in our study, it needs to be further investigated in future studies whether any beneficial associations between high HDL‐C levels and PBR_5‐9 as well as PWV still exist in levels well above HDL‐C > 100 mg/dL. The last limitation might be the use of the Sideview Dark Field imaging technique, which indirectly assess the EG thickness in the sublingual circulation, the latter may not be representative of all other microcirculatory beds. Although, the technique is operator‐dependent and requires substantial training due to a learning curve, it has important advantages like the noninvasive nature, the semi‐automated analyses, and the good reproducibility.

In conclusion, higher HDL‐C levels are associated with reduced aortic stiffness in the whole population while they protect EG and subsequently endothelial function in middle‐aged, treated hypertensive male patients independently of their smoking habit. We need more studies in several groups of low/high‐risk hypertensive patients in order to evaluate the beneficial role of elevated HDL‐C levels regarding CV risk as well as the significance of EG as a novel marker of target organ damage in essential hypertension.

CONFLICT OF INTEREST

The authors state that they have no conflict of interest to declare.

AUTHOR CONTRIBUTIONS

HT, DB, AS, MV, JT, GC, DB, and II substantially contributed to the conception or design of the work; or the acquisition, analysis, or interpretation of data for the work; and drafted the work or revising it critically for important intellectual content. HT and DB involved in final approval of the version to be published and agreed to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

Triantafyllidi H, Benas D, Schoinas A, et al. Sex‐related associations of high‐density lipoprotein cholesterol with aortic stiffness and endothelial glycocalyx integrity in treated hypertensive patients. J Clin Hypertens. 2020;22:1827–1834. 10.1111/jch.14002