Venous or Arterial Endothelium Evaluation for Early Cardiovascular Dysfunction in Hypertensive Patients?

Abstract

Veins and arteries have active endothelium, producing vasoactive substances like nitric oxide. The aim of this study was to evaluate whether hypertensive patients exhibit venous endothelial dysfunction and to determine the relationship between endothelial‐dependent and endothelial‐independent vasodilation responses in venous and arterial systems. Sixteen unmedicated patients with stage I and II hypertension and without other risk factors and 15 matched normotensive volunteers had venous and arterial endothelial function evaluated with the dorsal hand vein technique and brachial artery ultrasonography. Hypertensive patients had a marked reduction of maximum dilation to acetylcholine (54.9%±21.6%) compared with normotensive controls (85.2%±27.0%). The flow‐mediated dilation responses were reduced in hypertensive patients compared with controls (6.6%±3.3%vs 12.4%±2.6%, respectively). The responses to nitric oxide were similar in both groups, and the responses with the dorsal hand vein technique and flow‐mediated dilation agreed in both groups. Hypertensive patients had an attenuated endothelial dependent response, indicating that endothelial dysfunction is also present in the venous system.

Many studies have shown altered endothelial function in cardiovascular disorders. ¹ , ² , ³ Because the endothelium of conduit and resistance vessels is part of the target for vascular damage, early endothelial dysfunction in these conditions has been shown to result in atherosclerosis. ⁴ , ⁵ In this context, the loss of endothelial‐dependent vasodilation may be related to a reduction of nitric oxide (NO) bioavailability or activity. ⁶ , ⁷ , ⁸ , ⁹ , ¹⁰ , ¹¹ This impaired vasodilation can be measured in arterial and venous circulation by changes in vessel diameter as indexes of conduit vessel endothelial function. ¹² , ¹³ Otherwise, endothelial‐independent vascular function can be assessed by evaluating the vasodilation of an NO donor (high‐resolution ultrasonography). Although substantial evidence exists that endothelial‐dependent vasodilation is reduced in essential hypertension, ¹⁴ , ¹⁵ the degree of impairment of endothelial‐dependent vasodilation in the venous bed is unknown. Because veins are not susceptible to gross atherosclerosis, ¹⁶ high‐resolution ultrasonography's noninvasive approach can be used as a tool to assess vascular responsiveness in hypertensive patients. ¹⁷ , ¹⁸ This may be relevant when an increasing central blood volume, mediated in part by venous tone, appears to have a deleterious effect on cardiac remodeling and performance. ¹⁹

In the present study, we examined the extent of endothelial‐dependent and endothelial‐independent vasodilation in the venous and arterial beds in the same patient with essential hypertension, compared with a normotensive patient, and whether there is agreement in the response in these distinct vascular beds. The response could change the treatment approach to heart failure and other cases in which venous congestion could be the initial change occurring in cardiac failure in hypertensives. This kind of evaluation of endothelial‐dependent and endothelial independent vasodilation in both the venous and arterial beds might be useful in medical practice to follow these patients and their prognosis.

METHODS

This study was approved by the institution's ethics committee, and each participant provided written informed consent.

Study Population

The volunteers for this study were divided into 2 groups: (1) a normotensive group, which comprised 15 healthy patients with no familial history of coronary artery disease or arterial hypertension (blood pressure [BP] <140/90 mm Hg) and who were nonobese, nonhypercholesterolemic, nondiabetic, and nonsmokers; and (2) a hypertensive group, which comprised 16 stage I and II hypertensive patients (systolic BP within 140–179 mm Hg and diastolic BP 90–109 mm Hg) with no family history of coronary artery disease who were nonobese, nonhypercholesterolemic, and nondiabetic.

None of the patients were on regular medication, and all were clinically well. Patients with cardiac and vascular disease, impaired renal function, or other major pathologies were excluded from the study. In accordance with current legislation, all patients were aware of the investigational nature of the study and gave their written informed consent before participating.

Experimental Procedure

All studies were initiated at 8 am after an overnight fast, with the patients lying supine in a quiet air‐conditioned room (22°C–24°C). The patients were admitted to the institution on 2 occasions for a 4‐hour dorsal hand vein (DHV) technique study and a 2‐hour flow‐mediated dilation study described below.

DHV Technique

The DHV technique, previously modified by Aellig, ¹⁷ has been described in detail. Briefly, a 23‐gauge butterfly needle was inserted into a suitable vein on the back of the hand, with the arm positioned at an upward angle of 30° to allow the complete emptying of veins. A tripod, holding a linear variable differential transformer (LVDT) (Schaevitz Engineering, Pennsauken, NJ) was mounted on the back of the hand with the central aperture of the LVDT, containing a movable metal core, at a distance of 10 mm downstream from the tip of the needle. The signal output of the LVDT, which was linearly proportional to the vertical movement of the core, gave a measure of the diameter of the vein. Readings were taken under a congestive pressure of 40 mm Hg by inflating a BP cuff placed on the upper portion of the arm under study. Results are presented as normalized dose‐response curves in which the diameter of the vein during saline infusion is defined as 100% dilation. The vein was preconstricted to 20% of the baseline size by infusing increasing doses of phenylephrine (25–3166 ng/min). This dose rate of phenylephrine was defined as the ED₈₀ dose, and this degree of constriction was defined as 0% dilation for the purposes of subsequent calculations. The vasodilation effects expressed in this study were calculated as a percentage of the range between 0% and 100% dilation. Drugs were infused using a Harvard infusion pump (Harvard Apparatus Inc, South Natick, MA) at a flow rate of 0.3 mL/min. After preconstriction of the vein by using phenylephrine, a dose‐response curve of acetylcholine (0.36–3600 ng/min) and sodium nitroprusside (50–1000 ng/min) was constructed with 5 and 3 infusion doses in both normotensive and hypertensive groups, respectively. Systolic and diastolic BPs were measured with a mercury sphygmomanometer, and heart rate was measured by the pulse at the radial artery. All measurements were taken before and after each experimental phase.

Flow‐Mediated Dilation

Vascular responses in the brachial artery were studied by noninvasive high‐resolution ultrasonography (obtained with an ATL HDI system; Advanced Technology Laboratories, Seattle, WA), using a modification of the technique described by Celermajer and colleagues. ¹² , ²⁰ , ²¹ Arterial endothelial and vascular smooth muscle function were assessed by examining brachial artery responses to endothelial‐dependent (shear stress‐induced) and ‐independent (glyceryl trinitrate [GTN]‐mediated) stimuli. The patients rested quietly for 15 minutes before the first scan and remained in the supine position throughout the study.

The brachial artery was scanned longitudinally 5 to 10 cm above the elbow, and the center of the artery was identified when the clearest picture of the anterior and posterior arterial walls was obtained. When a satisfactory transducer position had been found, a special probe holder designed specifically for the study was fixed around the arm to secure the ultrasound transducer, and it was held in the same position throughout the study. Depth and gain settings were set to optimize the images of the lumen/arterial wall interface that were magnified using a resolution box function.

Arterial diameter measurements were performed at end‐diastole (R‐wave peak of the electrocardiogram) using electronic calipers. Four cardiac cycles were analyzed, and an average of the measurements was taken. The cycles were recorded on SuperVHS tape.

A baseline scan recorded the brachial artery diameter, and the arterial flow was measured using a pulsed Doppler signal at an angle of 60° relative to the wall of the artery with a 7‐MHz linear array transducer. Systolic and diastolic BPs were measured using a mercury sphygmomanometer, and heart rate was measured by feeling the radial artery pulse at the wrist for 1 minute. All measurements were taken before and after each experimental phase.

The shear stress‐induced vasodilator responses were used as a measure of endothelial‐dependent vasodilation. A pneumatic tourniquet was inflated around the arm to a pressure of 250 mm Hg for 4 to 5 minutes and then rapidly deflated. The resulting shear stress‐induced dilation (reactive hyperemia) in the arm and the increased brachial artery diameter were recorded from 15 to 90 seconds after cuff deflation. Changes in the brachial artery diameter in response to endothelial‐dependent NO‐mediated vasodilation induced by shear stress were expressed as a percentage change relative to the vessel diameter immediately before cuff inflation.

The brachial artery blood flow was calculated as the maximum flow recorded during the first 15 seconds after tourniquet release and was expressed as a percentage change relative to the flow immediately before cuff inflation. After allowing 10 to 15 minutes for brachial artery recovery, another baseline scan was performed.

The response to GTN was used as a measure of endothelial‐independent vasodilation. After recording the second baseline scan, 0.4 mg of GTN was given sublingually, and 4 minutes later the brachial artery was imaged. The response of the brachial artery diameter to GTN was expressed as a percentage change relative to the vessel diameter immediately before drug administration.

All images were analyzed by 2 observers who did not know the identity of the patients, the scan sequence, or the experimental phase.

Statistical Analyses

Descriptive data are expressed as the mean values ± SD. The maximum effect (Emax) with acetylcholine and sodium nitroprusside in the venous bed and the maximal vasodilation responses with reactive hyperemia and sublingual GTN in the artery were compared between normotensive and hypertensive patients. Parametric tests (t tests) were used to compare the maximal vasodilation values (Emax). The sample size was calculated for a power of .80. A P value <.05 was considered to indicate significance.

Agreement analysis between venous and arterial dilation of normotensive and hypertensive patients was performed according to the method of Bland and Altman. ²² The Bland‐Altman plot is a statistical method to compare 2 measurement techniques. In this graphical method the differences (or alternatively the ratios) between the 2 techniques are plotted against the averages of the 2 techniques. The graph displays a scatter diagram of the differences plotted against the averages of the 2 measurements. Horizontal lines are drawn at the mean difference and at the mean difference plus and minus 1.96 times the standard deviation of the differences. The agreement percentage is verified by data inside of the confidence interval (95%).

In the present study, we performed Bland‐Altman plots to analyze endothelial‐dependent and ‐independent dilation agreement between vein and artery territories, respectively, by comparing NO‐induced vasodilation responses produced by acetylcholine infusion vs shear stress and muscular relaxation induced by sodium nitroprusside infusion vs sublingual GTN.

RESULTS

Characteristics of the Groups

Table I summarizes the basic characteristics of the patients included in the study. There were no significant differences in age, body weight, body mass index, glycemia, total cholesterol and fractions, or triglyceride levels between the normotensive and hypertensive groups. No adverse effects were observed during the procedures.

Table I.

Clinical Characteristics of Normotensive and Hypertensive Patients

	Normotensive (n=15)	Hypertensive (n=16)
Age, y	37.4±4	40.0±6.4
Sex, male/female	6/9	9/7
Weight, kg	71.3±10	75.5±12
Body mass index, kg/m2	24.5±2	24.7±2
Glycemia, mg/dL	92.6±6	96.9±7
Cholesterol, mg/dL	195.8±34	201±32
HDL cholesterol, mg/dL	60.2±11	56.6±13
LDL cholesterol, mg/dL	116.9±30	121±29
Triglycerides, mg/dL	93.1±34	116.8±32
Resting systolic blood pressure, mm Hg	111.8±12	145.8±8.8a
Resting diastolic blood pressure, mm Hg	73±7	98.3±4.8a
Abbreviations: HDL, high‐density lipoprotein; LDL, low‐density lipoprotein. a P<.05 (t test).

Dorsal Hand Vein

Hypertensive patients had an impaired endothelial‐dependent venodilation to acetylcholine compared with normotensive patients (Emax, 54.9%±21.6% and 85.2%±27.0%, respectively; P<.05) (Table II). The endothelial‐nondependent venodilation response to sodium nitroprusside did not show significant differences (Table II).

Table II.

Endothelium‐Dependent and ‐Independent Responses in Veins and Arteries

	Dorsal Hand Vein		Flow‐Mediated Dilation
	Normotensive Group (n=15)	Hypertensive Group (n=16)	Normotensive Group (n=15)	Hypertensive Group (n=16)
Endothelium‐dependent dilation, %	85.2±27.0	54.9±21.6a	12.4±2.6	6.6±3.3a
Endothelium‐independent dilation, %	125.1±30	125±36	19.4±5	13.7±4
a P<.05 (t test).

Flow‐Mediated Dilation

Vasodilation induced by shear stress and GTN allowed the evaluation of the flow‐mediated dilation. The arterial response available through flow‐mediated dilation showed an impaired endothelial dependent vasodilation in hypertensive compared with normotensive patients (6.6%±3.3% vs 12.4%±2.6%, respectively; P<.05) (Table II). The endothelial‐independent vasodilation response was not significantly different between the groups (Table II).

Agreement Between DHV and Brachial Artery Responses

The endothelial‐dependent venodilation caused by acetylcholine and the arterial response through shear stress showed 80.0% and 62.0% agreement in the normotensive and hypertensive groups, respectively (Figure 1). Also the endothelial‐nondependent venodilation and the brachial arterial response showed 86.6% and 69.0% agreement in both groups, respectively (Figure 2). The arterial vasodilation values were adjusted (×10) to perform the Bland‐Altman plots.

Figure 1.

Agreement between endothelium‐dependent dilation in veins (in response to acetylcholine [Ach]) and arteries (in response to reactive hyperemia [flow‐mediated dilation (FMD)]) in normotensive patients (A) and hypertensive patients (B), based on Bland‐Altman calculations. The graph displays a scatter diagram of the differences plotted against the averages of the 2 methods (Ach and FMD). Horizontal lines are drawn at the mean difference and at the mean difference plus and minus 1.96 times the standard deviation of the differences. The agreement percentage, 80% in normotensive patients (A) and 62% in hypertensive patients (B), is verified by data inside of the confidence interval (95%).

Figure 2.

Agreement between endothelium‐independent dilation in veins (in response to sodium nitroprusside [SNP]) and arteries (in response to sublingual glyceryl trinitrate) in normotensive patients (A) and hypertensive patients (B), based on Bland‐Altman calculations. The graph displays a scatter diagram of the differences plotted against the average of the 2 methods (acetylcholine [Ach] and flow‐mediated dilation [FMD]). Horizontal lines are drawn at the mean difference and at the mean difference ± 1.96 times the standard deviation of the differences. The agreement percentage, 86.6% in normotensive patients (A) and 69% in hypertensive patients (B), is verified by data inside of the confidence interval (95%).

Hemodynamics

All measurements of systolic and diastolic BP and heart rate were taken before and after each experimental phase during the DHV and the flow‐mediated dilation. Infusion of acetylcholine and sodium nitroprusside did not cause systemic changes (data not shown).

Correlation Between Glycemia and Vascular Responses

Fasting plasma glucose levels and endothelium‐dependent venodilation had a significant inverse correlation in hypertensive patients (R=−0.50; P<.05) (Figure 3). However, we are not able to correlate arterial reactivity and glycemia levels with this design of study.

Figure 3.

Correlation between fasting glycemia plasma levels and venous endothelium‐dependent dilation in normotensive and hypertensive groups.

DISCUSSION

First, the results of this observational study show that large veins (superficial DHV) and arteries (brachial artery) showed a significant impairment in dilation mediated by the endothelium in hypertensive patients. In contrast, endothelium‐independent vasodilation was not altered in the same patients. Although some vascular relaxing effects of NO on arteries and veins has been reported, ²³ , ²⁴ , ²⁵ to the best of our knowledge their endothelial‐dependent and ‐independent responses have never been put forward and have not been evaluated in the same hypertensive patients. The inverse correlation between normal glucose levels and vascular function in human veins and arteries constitutes another interesting finding, which deserves further study. Any conclusions from these data would be speculative.

Arterial hypertension is a complex disease in which many pathophysiologic factors play a role in association with endothelial dysfunction. Hemodynamics, sympathetic activity, renin‐angiotensin system activation, production of derived‐endothelial vasoactive substances (endothelin, prostaglandins, thromboxane), and the balance of radical free oxygen species interact to maintain the vascular tone. NO has been considered the major substance responsible for the relaxation of the endothelium. ⁸ In contrast with the continuous production of NO by arterial endothelium, the basal production of NO from venous endothelium is low. ²⁶ It has been demonstrated that endothelial‐dependent relaxation is lower in veins than that in arteries, although the vasodilation caused by physiologic release of NO is similar in both vessels. ²³ Some studies indicate that release and production of NO from vein endothelium can be increased in response to acetylcholine or other molecules. ²⁷ Other studies suggest, however, that the venous endothelium has a lower capacity for NO synthesis than does arterial endothelium, and in some cases may generate less NO. For example, cultured endothelial cells from bovine aorta and vena cava release similar amounts of NO in response to bradykinin, ²⁸ shear stress and acetylcholine stimulate less NO release from the jugular vein than from the nearby carotid artery, ²⁹ and shear stress elicits a smaller NO‐dependent dilation in coronary venules than in their paired arterioles. ³⁰ In vitro studies comparing the internal mammary artery with the saphenous vein in humans reinforced the effect of endothelial‐dependent relaxation to acetylcholine in both vessels, although a more marked response has been observed in the artery than in the vein. ³¹ Studies have reported the superior long‐term patency of internal mammary artery grafts over saphenous vein grafts based on endothelial function. A higher basal release of NO by arterial grafts, ³² with intact endothelial function and absence of intimal thickening in comparison with saphenous grafts, ³³ could explain the presence of physiologic NO and prostacyclin, in arterial more than in venous grafts, ²³ suggesting a higher bioavailability of NO in arterial beds. In vivo, an NO biosynthesis inhibitor (N^G‐monometil‐L‐arginine) unfunded in superficial DHVs confirmed the findings described previously for arteries. For both vessels, the authors observed normal maximum vasodilator dilation to an NO donor, nitroglycerin, characterizing the responses as endothelial‐dependent and ‐independent, respectively. ²⁶

Panza and colleagues ³⁴ were the first to report that the endothelium in hypertensive patients has a reduced dilation to acetylcholine compared with that in normal patients. Later studies involving patients genetically predisposed to developing arterial hypertension demonstrated endothelial dysfunction, which was associated with reduced NO bioactivity. ³⁵ , ³⁶ Widgren and colleagues ³⁷ also found that normotensive patients with a family history of hypertension display decreased venous compliance and an increased arterial pressure response to acute increases in vascular fluid volume. These acute findings reinforce impaired vascular dysfunction endothelial‐mediated in both veins and arteries in the hypertensive disease we found; the responsiveness to NO donors is still controversial. ⁷ , ³⁸ Furthermore, a close agreement between the responses of veins and arteries was observed in both normotensive and hypertensive groups in this study.

Although the link between diabetes and dysglycemia conditions to vascular disease is not fully understood, loss of the modulatory role of the endothelium is implicated in the pathogenesis of diabetic microangiopathy and macroangiopathy. ³⁹ , ⁴⁰ On the other hand, disturbances in glucose metabolism result in oxidative stress and the increased formation of advanced glycosylation end products, both of which increase the inactivation of NO. ⁴¹ Also, advanced glycation end product accumulation on wall vessels causes stiffening of the collagen fibers thereby affecting arterial mechanical properties. ⁴² From our data, we can hypothesize that a normal but increasing range of blood glucose levels could be related to different degrees of endothelial dysfunction.

Study Limitations

The functional effects of NO on systemic circulation are not restricted to the large vessels. It is well‐known that arterioles are primarily responsible for total systemic arterial resistance, and the methods used in our study are not adequate for this evaluation. Therefore, we did not analyze all the functional repercussions that can explain the early pathophysiology of vascular dysfunction (mediated or not by the endothelium) in hypertensive patients.

Also, the inverse correlation between glucose blood levels and vascular function in human veins cannot be extrapolated to arteries for diabetic or prediabetic status.

CONCLUSIONS

Our results demonstrated an impaired responsiveness in arterial and venous endothelial‐dependent vascular reactivity in hypertensive patients and an agreement of responses between these 2 vessel beds to acetylcholine and nitroglycerin. These results may explain some vascular adaptations (volume capacitance) and early atherogenesis. The inverse correlation between normal glucose levels and vascular function deserves further investigation; diabetic and prediabetic states play an important role in vascular reactivity in human veins and arteries.

Disclosure:

This study was supported by Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP, Brazil) and Conselho Nacional de Pesquisa (CNPq, Brazil).